作者Andrei Chetvergov, Alexander Evseev, Mikhail Solovev, Timofei Sivoraksha, Stepan Ukolov, Valeriia Kuschenko, Maria Chistyakova, Sergey Bolovtsov
Large language models trained and aligned within different linguistic and regional ecosystems may frame the same political, cultural, and geopolitical entities in different ways. Such differences are often evaluated through sentiment, favorability, or stance, reducing model attitudes to a single positive-negative axis. We introduce REGARD, a study of what drives affective framing differences across LLMs on post-Soviet entities using target-directed Valence-Arousal-Dominance profiling. We query 19 models on 500 region-specific targets, score their responses with two independent LLM judges, GPT-4o-mini and Qwen3.6-35B-A3B, and validate the measurements on a 300-item human-annotated subset. Post-hoc Ward-linkage clustering of all 19 models by affective and response-behavior profiles yields three behavioral clusters that cut across model origin, family, and parameter count. Generic-answer rate is strongly associated with lower arousal (r = -0.81) and with cluster placement: models that deflect evaluative prompts with templated responses cluster together at low arousal regardless of origin. These findings show that VAD profiling captures emotional intensity, a dimension of affective framing that is largely invisible to conventional sentiment-based evaluation.
展开完整摘要收起摘要↓
Large language models trained and aligned within different linguistic and regional ecosystems may frame the same political, cultural, and geopolitical entities in different ways. Such differences are often evaluated through sentiment, favorability, or stance, reducing model attitudes to a single positive-negative axis. We introduce REGARD, a study of what drives affective framing differences across LLMs on post-Soviet entities using target-directed Valence-Arousal-Dominance profiling. We query 19 models on 500 region-specific targets, score their responses with two independent LLM judges, GPT-4o-mini and Qwen3.6-35B-A3B, and validate the measurements on a 300-item human-annotated subset. Post-hoc Ward-linkage clustering of all 19 models by affective and response-behavior profiles yields three behavioral clusters that cut across model origin, family, and parameter count. Generic-answer rate is strongly associated with lower arousal (r = -0.81) and with cluster placement: models that deflect evaluative prompts with templated responses cluster together at low arousal regardless of origin. These findings show that VAD profiling captures emotional intensity, a dimension of affective framing that is largely invisible to conventional sentiment-based evaluation.
Coordination, a fundamental linguistic structure, remains a subject of intense debate, and its exact nature continues to elude theoretical linguistics. A common view holds that only same-category constituents can be conjoined, which has been challenged by the many grammatical unlike coordinations found in natural language. Treating language models as a computational testbed, we investigate whether the acquisition of unlike coordination requires direct exposure in the training data, or whether it can emerge organically from general compositional abilities. Using Filtered-Corpus Training (FiCT), we train GPT-2 models on corpora from which all instances of unlike coordination have been removed. We find that direct exposure is not necessary: models trained on filtered data successfully generalize to unlike coordination, achieving perplexity and grammaticality judgments comparable to models trained on unfiltered text. Furthermore, our analyses of internal representations indicate that language models process unlike coordination by treating the conjoined elements as belonging to similar structural categories or through a mechanism akin to deletion, both of which appear learnable from exposure to alike coordination alone. This work contributes to the growing understanding of how language models internally represent linguistic structures, while also adding to the broader debate on coordination by showing how models generalize and process unlike coordination without direct exposure.
展开完整摘要收起摘要↓
Coordination, a fundamental linguistic structure, remains a subject of intense debate, and its exact nature continues to elude theoretical linguistics. A common view holds that only same-category constituents can be conjoined, which has been challenged by the many grammatical unlike coordinations found in natural language. Treating language models as a computational testbed, we investigate whether the acquisition of unlike coordination requires direct exposure in the training data, or whether it can emerge organically from general compositional abilities. Using Filtered-Corpus Training (FiCT), we train GPT-2 models on corpora from which all instances of unlike coordination have been removed. We find that direct exposure is not necessary: models trained on filtered data successfully generalize to unlike coordination, achieving perplexity and grammaticality judgments comparable to models trained on unfiltered text. Furthermore, our analyses of internal representations indicate that language models process unlike coordination by treating the conjoined elements as belonging to similar structural categories or through a mechanism akin to deletion, both of which appear learnable from exposure to alike coordination alone. This work contributes to the growing understanding of how language models internally represent linguistic structures, while also adding to the broader debate on coordination by showing how models generalize and process unlike coordination without direct exposure.
作者Andrei Chetvergov, Stepan Ukolov, Timofei Sivoraksha, Alexander Evseev, Mikhail Solovev, Valeriia Kuschenko, Maria Chistyakova, Sergey Bolovtsov
Large language models are increasingly evaluated through the values they endorse, but such evaluations presuppose that models can identify the value expressed in a concrete situation. We study this prerequisite as controlled top-1 recognition over Schwartz's ten basic values. Our evaluation set contains 1,000 Russian situational texts, balanced across the ten values and independently labeled by two human annotators per item. We evaluate 21 instruction-tuned LLM runs under a fixed ranked-response protocol; 20 runs with reliable outputs form the semantic panel. Pooled Acc@1 is 0.683 and Acc@3 is 0.892, showing that models often locate the correct motivational region while ranking close alternatives unstably. Adjacent values account for 50.9% of semantic errors, compared with 24.4% under a checkpoint-specific null. Eight directed confusions recur across checkpoints and human-confirmed subsets. Several are strongly asymmetric, including Universalism to Benevolence, Tradition to Conformity, and Security to Power, whereas Stimulation-Hedonism forms a bidirectional boundary. Their severity is checkpoint-specific and can bias higher-order value profiles. The results motivate value-recognition evaluation that combines exact accuracy, ranked recovery, and directed error analysis.
展开完整摘要收起摘要↓
Large language models are increasingly evaluated through the values they endorse, but such evaluations presuppose that models can identify the value expressed in a concrete situation. We study this prerequisite as controlled top-1 recognition over Schwartz's ten basic values. Our evaluation set contains 1,000 Russian situational texts, balanced across the ten values and independently labeled by two human annotators per item. We evaluate 21 instruction-tuned LLM runs under a fixed ranked-response protocol; 20 runs with reliable outputs form the semantic panel. Pooled Acc@1 is 0.683 and Acc@3 is 0.892, showing that models often locate the correct motivational region while ranking close alternatives unstably. Adjacent values account for 50.9% of semantic errors, compared with 24.4% under a checkpoint-specific null. Eight directed confusions recur across checkpoints and human-confirmed subsets. Several are strongly asymmetric, including Universalism to Benevolence, Tradition to Conformity, and Security to Power, whereas Stimulation-Hedonism forms a bidirectional boundary. Their severity is checkpoint-specific and can bias higher-order value profiles. The results motivate value-recognition evaluation that combines exact accuracy, ranked recovery, and directed error analysis.
Activation steering provides a lightweight mechanism for controlling large language models at inference time, but steering vectors can have unintended externalities: utility vectors may weaken safety behavior, while refusal vectors may induce over-refusal on benign prompts. We introduce OPIUM (Optimizing Protected Injections via Utility Manifolds), a training-free method for sanitizing steering vectors through representation matching. Given reference behaviors on two prompt sets, OPIUM optimizes a new steering vector that preserves the downstream representations induced by the desired intervention while matching a safer reference behavior on prompts where the original vector fails. Across steering-externality and over-refusal settings, OPIUM improves the safety--utility tradeoff relative to vanilla steering and directional ablation, suggesting that harmful side effects of activation steering can often be mitigated directly in activation space.
展开完整摘要收起摘要↓
Activation steering provides a lightweight mechanism for controlling large language models at inference time, but steering vectors can have unintended externalities: utility vectors may weaken safety behavior, while refusal vectors may induce over-refusal on benign prompts. We introduce OPIUM (Optimizing Protected Injections via Utility Manifolds), a training-free method for sanitizing steering vectors through representation matching. Given reference behaviors on two prompt sets, OPIUM optimizes a new steering vector that preserves the downstream representations induced by the desired intervention while matching a safer reference behavior on prompts where the original vector fails. Across steering-externality and over-refusal settings, OPIUM improves the safety--utility tradeoff relative to vanilla steering and directional ablation, suggesting that harmful side effects of activation steering can often be mitigated directly in activation space.
作者Abigail Woodring, Adrian Chan, Rana Muhammad Shahroz Khan, Sukwon Yun, Chau-Wai Wong, Tianlong Chen
Fine-tuning has been widely used to adapt large language models (LLMs) for domain-specific tasks. Parameter efficient fine-tuning (PEFT) methods such as low-rank adaptation (LoRA) are frequently used to reduce computational costs. PortLLM is a training-free and data-free scheme used to adapt LLMs after continual pretraining. Although the initial PortLLM results show that LoRA patches exhibit short-term temporal portability, the long-term performance of PortLLM across several updates of continual pretraining remains underexplored. Furthermore, the intriguing effectiveness of PortLLM is not well understood from a theoretical standpoint. We address these two open questions by (1) performing an extensive empirical study of the long-term temporal portability of PortLLM patches across 10 continual pretraining steps using base models Mistral, Gemma, and Qwen; and (2) offering two theoretical analyses to explain our observation that the simple PortLLM method achieves competitive performance. We find empirically that the portability persists across longer time duration, indicating that repeated fine-tuning is not required when the base model is periodically updated. We find theoretically that near-orthogonality of high-dimensional vectors is a key justification for temporal portability. Our analyses also demonstrate a geometric perspective of the loss landscape in facilitating the theoretical comparison of different adaptation options.
展开完整摘要收起摘要↓
Fine-tuning has been widely used to adapt large language models (LLMs) for domain-specific tasks. Parameter efficient fine-tuning (PEFT) methods such as low-rank adaptation (LoRA) are frequently used to reduce computational costs. PortLLM is a training-free and data-free scheme used to adapt LLMs after continual pretraining. Although the initial PortLLM results show that LoRA patches exhibit short-term temporal portability, the long-term performance of PortLLM across several updates of continual pretraining remains underexplored. Furthermore, the intriguing effectiveness of PortLLM is not well understood from a theoretical standpoint. We address these two open questions by (1) performing an extensive empirical study of the long-term temporal portability of PortLLM patches across 10 continual pretraining steps using base models Mistral, Gemma, and Qwen; and (2) offering two theoretical analyses to explain our observation that the simple PortLLM method achieves competitive performance. We find empirically that the portability persists across longer time duration, indicating that repeated fine-tuning is not required when the base model is periodically updated. We find theoretically that near-orthogonality of high-dimensional vectors is a key justification for temporal portability. Our analyses also demonstrate a geometric perspective of the loss landscape in facilitating the theoretical comparison of different adaptation options.
Aspect-based sentiment analysis (ABSA) in Arabic must recover both explicitly stated aspects and implicit aspects that are never named in the text. Implicit identification typically relies on an auxiliary knowledge source (e.g., a knowledge graph (KG)) linking opinion cues to aspect categories, but for a lower-resource language the practitioner faces a design choice: reuse a mature English KG through multilingual embeddings, or build a smaller native Arabic KG. This paper reports a controlled comparison of the two strategies within a single hybrid pipeline, evaluated on three Arabic benchmarks (M-ABSA, SemEval-2016 Arabic, and HAAD). We further compare two adaptation strategies for the generative extractor that feeds the KG — zero-shot prompting versus task-specific fine-tuning of an 8B-parameter large language model (LLM). The native Arabic KG (Strategy 2) outperforms the cross-lingual English KG (Strategy 1) by +0.199 micro-F1 on M-ABSA and +0.251 on SemEval-2016, gaining on both precision and recall. Task-specific fine-tuning raises explicit-extraction micro-F1 from <= 0.13 (zero-shot) to 0.66-0.76 on M-ABSA and SemEval-2016 (0.45 on the smaller HAAD), confirming that task adaptation, rather than model scale, is decisive in a morphologically rich language.
展开完整摘要收起摘要↓
Aspect-based sentiment analysis (ABSA) in Arabic must recover both explicitly stated aspects and implicit aspects that are never named in the text. Implicit identification typically relies on an auxiliary knowledge source (e.g., a knowledge graph (KG)) linking opinion cues to aspect categories, but for a lower-resource language the practitioner faces a design choice: reuse a mature English KG through multilingual embeddings, or build a smaller native Arabic KG. This paper reports a controlled comparison of the two strategies within a single hybrid pipeline, evaluated on three Arabic benchmarks (M-ABSA, SemEval-2016 Arabic, and HAAD). We further compare two adaptation strategies for the generative extractor that feeds the KG — zero-shot prompting versus task-specific fine-tuning of an 8B-parameter large language model (LLM). The native Arabic KG (Strategy 2) outperforms the cross-lingual English KG (Strategy 1) by +0.199 micro-F1 on M-ABSA and +0.251 on SemEval-2016, gaining on both precision and recall. Task-specific fine-tuning raises explicit-extraction micro-F1 from <= 0.13 (zero-shot) to 0.66-0.76 on M-ABSA and SemEval-2016 (0.45 on the smaller HAAD), confirming that task adaptation, rather than model scale, is decisive in a morphologically rich language.
作者Sriprabha Ramanarayanan, Rahul G. S., Mohammad Al Fahim, Keerthi Ram, Ramesh Venkatesan, Mohanasankar Sivaprakasam
Attention Mechanism (AM) selectively focuses on essential information for imaging tasks and captures relationships between distant pixel neighborhoods to compute feature representations. Accelerated MRI reconstruction benefits from AM, as the imaging process involves Fourier domain measurements that influence image representation non-locally. However, AM-based models are more adept at capturing low-frequency information with limited capacity for high-frequency representations, restricting models to smooth reconstruction. Additionally, AM-based models need mode-specific retraining for multimodal MRI data, as their knowledge is restricted to local contextual variations that may be inadequate to capture transferable features across heterogeneous domains. To address these challenges, we propose a neuromodulation-based discriminative multi-spectral AM for scalable MRI reconstruction that can (i) propagate context-aware high-frequency details for high-quality reconstruction, and (ii) capture features reusable across deviated unseen domains in multimodal MRI. The proposed network consists of a spectral filtering CNN to capture mode-specific transferable features and a dynamic high-pass kernel generation transformer focusing on high-frequency details. We evaluate our model on comparative studies in supervised and self-supervised learning, diffusion model-based training, closed-set and open-set generalization under heterogeneous MRI data, and interpretation-based analysis. Our method offers scalable, high-quality reconstruction with best improvement margins of ~1 dB in PSNR and ~0.01 in SSIM under unseen scenarios. Code: https://github.com/sriprabhar/SHFormer
展开完整摘要收起摘要↓
Attention Mechanism (AM) selectively focuses on essential information for imaging tasks and captures relationships between distant pixel neighborhoods to compute feature representations. Accelerated MRI reconstruction benefits from AM, as the imaging process involves Fourier domain measurements that influence image representation non-locally. However, AM-based models are more adept at capturing low-frequency information with limited capacity for high-frequency representations, restricting models to smooth reconstruction. Additionally, AM-based models need mode-specific retraining for multimodal MRI data, as their knowledge is restricted to local contextual variations that may be inadequate to capture transferable features across heterogeneous domains. To address these challenges, we propose a neuromodulation-based discriminative multi-spectral AM for scalable MRI reconstruction that can (i) propagate context-aware high-frequency details for high-quality reconstruction, and (ii) capture features reusable across deviated unseen domains in multimodal MRI. The proposed network consists of a spectral filtering CNN to capture mode-specific transferable features and a dynamic high-pass kernel generation transformer focusing on high-frequency details. We evaluate our model on comparative studies in supervised and self-supervised learning, diffusion model-based training, closed-set and open-set generalization under heterogeneous MRI data, and interpretation-based analysis. Our method offers scalable, high-quality reconstruction with best improvement margins of ~1 dB in PSNR and ~0.01 in SSIM under unseen scenarios. Code: https://github.com/sriprabhar/SHFormer
Large Audio Language models (LALMs) have made rapid progress on acoustic understanding, yet they still struggle with fine-grained audio reasoning (e.g., recognizing event order, repetitions and duration). Existing post-training methods heavily rely on expensive external labels or provide only coarse semantic signals. To bridge this gap, we introduce Audio-Zero, the first label-free self-evolution framework in the field of LALMs that improves fine-grained auditory perception and reasoning. Audio-Zero constructs an auditory self-play game from unlabeled audio contrast pairs: most players hear a reference audio, while one odd listener hears a subtle variant. The model first generates clues describing what it hears and then identifies the odd listener by reasoning over inconsistencies among clues. Since the odd listener is known by construction, the game provides verifiable rewards without any annotated answers. Experiments with Qwen2-Audio-7B-Instruct and Qwen2.5-Omni-7B on TREA, MMAU Test-mini and MMAR show that Audio-Zero improves fine-grained audio reasoning while preserving broad audio understanding. Evolutionary and diagnostic analyses further reveal that increasingly fine-grained auditory descriptions emerge naturally from game pressure.
展开完整摘要收起摘要↓
Large Audio Language models (LALMs) have made rapid progress on acoustic understanding, yet they still struggle with fine-grained audio reasoning (e.g., recognizing event order, repetitions and duration). Existing post-training methods heavily rely on expensive external labels or provide only coarse semantic signals. To bridge this gap, we introduce Audio-Zero, the first label-free self-evolution framework in the field of LALMs that improves fine-grained auditory perception and reasoning. Audio-Zero constructs an auditory self-play game from unlabeled audio contrast pairs: most players hear a reference audio, while one odd listener hears a subtle variant. The model first generates clues describing what it hears and then identifies the odd listener by reasoning over inconsistencies among clues. Since the odd listener is known by construction, the game provides verifiable rewards without any annotated answers. Experiments with Qwen2-Audio-7B-Instruct and Qwen2.5-Omni-7B on TREA, MMAU Test-mini and MMAR show that Audio-Zero improves fine-grained audio reasoning while preserving broad audio understanding. Evolutionary and diagnostic analyses further reveal that increasingly fine-grained auditory descriptions emerge naturally from game pressure.
作者Yurong Liu, Yeye He, Haoyu Dong, Junjie Xing, Shi Han, Dongmei Zhang, Surajit Chaudhuri
Predicting missing cell values in tabular data is a fundamental problem in data cleaning. While state-of-the-art reasoning models show great promise in predicting missing values in tables, by reasoning holistically across rows and columns, they are costly to deploy at scale and tend to be overconfident, often generating hallucinated or false-positive predictions. In this paper, we observe that achieving high-precision missing-value prediction in tables requires a distinct combination of three capabilities: (1) world knowledge, (2) text-based reasoning, and (3) code-based reasoning. We systematically explore design choices for combining these capabilities, and propose an Auto-Fill approach that post-trains three specialist small language models (SLMs), each optimized for one capability. We develop a calibrated ensemble mechanism that either dynamically selects the most confident specialist or abstains, ensuring high accuracy. Extensive experiments on 11 benchmarks with 2200 real tables drawn from diverse domains show that Auto-Fill achieves superior accuracy compared to state-of-the-art reasoning models (e.g., o3-pro, Gemini 3 Pro, and DeepSeek R1), while operating at a fraction (less than 1%) of the cost of these frontier models. Our results highlight the effectiveness of specialization and calibrated abstention in the important domain of tabular data. Auto-Fill is publicly available at https://github.com/lyrain2001/auto-fill.
展开完整摘要收起摘要↓
Predicting missing cell values in tabular data is a fundamental problem in data cleaning. While state-of-the-art reasoning models show great promise in predicting missing values in tables, by reasoning holistically across rows and columns, they are costly to deploy at scale and tend to be overconfident, often generating hallucinated or false-positive predictions. In this paper, we observe that achieving high-precision missing-value prediction in tables requires a distinct combination of three capabilities: (1) world knowledge, (2) text-based reasoning, and (3) code-based reasoning. We systematically explore design choices for combining these capabilities, and propose an Auto-Fill approach that post-trains three specialist small language models (SLMs), each optimized for one capability. We develop a calibrated ensemble mechanism that either dynamically selects the most confident specialist or abstains, ensuring high accuracy. Extensive experiments on 11 benchmarks with 2200 real tables drawn from diverse domains show that Auto-Fill achieves superior accuracy compared to state-of-the-art reasoning models (e.g., o3-pro, Gemini 3 Pro, and DeepSeek R1), while operating at a fraction (less than 1%) of the cost of these frontier models. Our results highlight the effectiveness of specialization and calibrated abstention in the important domain of tabular data. Auto-Fill is publicly available at https://github.com/lyrain2001/auto-fill.
Low-rank adaptation (LoRA) has become a widely used parameter-efficient fine-tuning method for large language models. Since different modules and layers may contribute unequally to downstream adaptation, allocating rank resources under a fixed parameter budget is an important problem for balancing efficiency, expressiveness, and generalization. Existing adaptive rank methods address this problem mainly through carefully designed importance scores constructed from gradient-derived sensitivity and uncertainty measures, without an explicit statistical interpretation. In this paper, we formulate LoRA rank allocation as a statistical hypothesis testing problem and propose StatLoRA, a statistical inference-based rank allocation method. StatLoRA associates each LoRA component with a test statistic and uses estimated p-values to determine which components should be retained or pruned under a prescribed rank budget. The proposed testing procedure is supported by our central limit theory for stochastic optimizer trajectories. In particular, we establish asymptotic normality for a broad class of commonly used optimizers in deep learning, including AdamW, and derive the corresponding asymptotic distributions for the proposed component scores used in hypothesis testing. We evaluate StatLoRA on LoRA fine-tuning of DeBERTaV3-base, BART-Large, and Qwen2.5-7B across natural language understanding, natural language generation, and question answering tasks. Experiments show that StatLoRA achieves comparable or better performance than vanilla LoRA, AdaLoRA, and IGU-LoRA under matched rank budgets. Sensitivity analyses and empirical diagnostics further support the stability of the proposed hypothesis-testing-based allocation rule and provide empirical evidence for the asymptotic theory of component scores.
展开完整摘要收起摘要↓
Low-rank adaptation (LoRA) has become a widely used parameter-efficient fine-tuning method for large language models. Since different modules and layers may contribute unequally to downstream adaptation, allocating rank resources under a fixed parameter budget is an important problem for balancing efficiency, expressiveness, and generalization. Existing adaptive rank methods address this problem mainly through carefully designed importance scores constructed from gradient-derived sensitivity and uncertainty measures, without an explicit statistical interpretation. In this paper, we formulate LoRA rank allocation as a statistical hypothesis testing problem and propose StatLoRA, a statistical inference-based rank allocation method. StatLoRA associates each LoRA component with a test statistic and uses estimated p-values to determine which components should be retained or pruned under a prescribed rank budget. The proposed testing procedure is supported by our central limit theory for stochastic optimizer trajectories. In particular, we establish asymptotic normality for a broad class of commonly used optimizers in deep learning, including AdamW, and derive the corresponding asymptotic distributions for the proposed component scores used in hypothesis testing. We evaluate StatLoRA on LoRA fine-tuning of DeBERTaV3-base, BART-Large, and Qwen2.5-7B across natural language understanding, natural language generation, and question answering tasks. Experiments show that StatLoRA achieves comparable or better performance than vanilla LoRA, AdaLoRA, and IGU-LoRA under matched rank budgets. Sensitivity analyses and empirical diagnostics further support the stability of the proposed hypothesis-testing-based allocation rule and provide empirical evidence for the asymptotic theory of component scores.
作者Kabir Murjani, Mishri Bhavsar, Manish I. Patel, Jonti Talukdar
Very Large Scale Integration (VLSI) global routing is an NP-hard combinatorial optimization problem requiring signal net assignment across capacity-constrained 3D grids while minimizing congestion, wirelength, and via transitions. Because traditional heuristics rely on static penalty schedules that fail on complex congestion topologies, we present AlphaRoute: a multi-objective adaptive search framework reformulating rip-up and reroute (R&R) into a dynamic optimization system. We introduce SHAP-based overflow decomposition to isolate per-net congestion, driving targeted subgraph extraction via 3D Dijkstra maze routing and an adaptive PathFinder policy. Crucially, AlphaRoute employs Large Language Models (LLMs) as semantic policy optimizers. Bounded by a deterministic knowledge graph, the LLMs interpret congestion metrics to dynamically adjust penalty parameters. Evaluated on ISPD 2025 benchmarks, AlphaRoute reduces overflow by 98.6% on MEMPOOL. On the constrained ARIANE design, we achieve an overflow of 146,109 (a 29.8x reduction in overflow over the state of the art), yielding a penalized score of S_orig = 0.0538 versus the State-of-the-art (SOTA) 1.780. These results demonstrate that superior algorithmic search geometry can overcome the latency of interpreted Python implementations.
展开完整摘要收起摘要↓
Very Large Scale Integration (VLSI) global routing is an NP-hard combinatorial optimization problem requiring signal net assignment across capacity-constrained 3D grids while minimizing congestion, wirelength, and via transitions. Because traditional heuristics rely on static penalty schedules that fail on complex congestion topologies, we present AlphaRoute: a multi-objective adaptive search framework reformulating rip-up and reroute (R&R) into a dynamic optimization system. We introduce SHAP-based overflow decomposition to isolate per-net congestion, driving targeted subgraph extraction via 3D Dijkstra maze routing and an adaptive PathFinder policy. Crucially, AlphaRoute employs Large Language Models (LLMs) as semantic policy optimizers. Bounded by a deterministic knowledge graph, the LLMs interpret congestion metrics to dynamically adjust penalty parameters. Evaluated on ISPD 2025 benchmarks, AlphaRoute reduces overflow by 98.6% on MEMPOOL. On the constrained ARIANE design, we achieve an overflow of 146,109 (a 29.8x reduction in overflow over the state of the art), yielding a penalized score of S_orig = 0.0538 versus the State-of-the-art (SOTA) 1.780. These results demonstrate that superior algorithmic search geometry can overcome the latency of interpreted Python implementations.
Quantized small autoregressive reasoning models can enter long, repetitive, or unproductive trajectories, yet inference-time compute is usually allocated without observing how a trajectory develops. Building on an earlier token-level e-CUSUM controller, we develop MGT-B (Monitoring-Guided Test-time Backtracking), a revised external controller that maps overlapping windows of pre-sampling uncertainty and degeneration features to position-conditional empirical tail probabilities, accumulates mixture betting factors with a CUSUM-shaped reset, and responds to an alarm by estimating a rollback point, restoring token and key-value-cache state, and performing constrained re-decoding. To audit whether the effect persists on problem identities first observed after the manual choice of log threshold h = 10, we retrospectively exclude 260 IDs present in pre-threshold artifacts and retain the chronologically first post-threshold pair for each remaining ID, yielding a 240-pair chronology-audit set. On this set, accuracy changes from 82/240 to 88/240 (+2.50 percentage points; 13 corrections, 7 regressions; exact McNemar p = 0.2632; paired bootstrap 95% interval [-1.25, +6.25]). A broader 467-pair historical-coverage set of seed-matched pairs changes accuracy from 146/467 to 167/467 (+4.50 points; McNemar p = 0.000753), but includes 200 seed-1 IDs available before or during threshold selection and is reported only as an exploratory estimate. All 316 no-alarm outputs in the 467-pair set are identical to vanilla, while the 151 alarmed trajectories contain 29 corrections and 8 regressions. Neither analysis is confirmatory, and the empirical factors are not established as a valid e-process or e-detector. The results support a selective monitoring-and-repair mechanism for the studied MATH-500 setting, rather than a general or theoretically certified reasoning improvement.
展开完整摘要收起摘要↓
Quantized small autoregressive reasoning models can enter long, repetitive, or unproductive trajectories, yet inference-time compute is usually allocated without observing how a trajectory develops. Building on an earlier token-level e-CUSUM controller, we develop MGT-B (Monitoring-Guided Test-time Backtracking), a revised external controller that maps overlapping windows of pre-sampling uncertainty and degeneration features to position-conditional empirical tail probabilities, accumulates mixture betting factors with a CUSUM-shaped reset, and responds to an alarm by estimating a rollback point, restoring token and key-value-cache state, and performing constrained re-decoding. To audit whether the effect persists on problem identities first observed after the manual choice of log threshold h = 10, we retrospectively exclude 260 IDs present in pre-threshold artifacts and retain the chronologically first post-threshold pair for each remaining ID, yielding a 240-pair chronology-audit set. On this set, accuracy changes from 82/240 to 88/240 (+2.50 percentage points; 13 corrections, 7 regressions; exact McNemar p = 0.2632; paired bootstrap 95% interval [-1.25, +6.25]). A broader 467-pair historical-coverage set of seed-matched pairs changes accuracy from 146/467 to 167/467 (+4.50 points; McNemar p = 0.000753), but includes 200 seed-1 IDs available before or during threshold selection and is reported only as an exploratory estimate. All 316 no-alarm outputs in the 467-pair set are identical to vanilla, while the 151 alarmed trajectories contain 29 corrections and 8 regressions. Neither analysis is confirmatory, and the empirical factors are not established as a valid e-process or e-detector. The results support a selective monitoring-and-repair mechanism for the studied MATH-500 setting, rather than a general or theoretically certified reasoning improvement.
We propose a novel framework for computing rigorous bounds on the probability that a large language model (LLM) generates harmful output to a given prompt. We study a new application of the Clopper-Pearson confidence intervals to obtain probably approximately correct (PAC) bounds for this problem. As our main technical contribution, we propose an algorithm that leverages features in the latent space to prioritize exploring branches in the auto-regressive generation tree that are more likely to produce harmful outputs. Our approach in particular enables the efficient computation of useful lower bounds, even in scenarios where the true harm probability is extremely small, and crucially, the obtained lower bounds are sound, i.e., formally proven to be less than the actual harmfulness probability: our experimental results demonstrate the effectiveness of our method by computing non-trivial lower bounds on state-of-the-art LLMs. This study newly enables the evaluation and statistical certification of LLMs.
展开完整摘要收起摘要↓
We propose a novel framework for computing rigorous bounds on the probability that a large language model (LLM) generates harmful output to a given prompt. We study a new application of the Clopper-Pearson confidence intervals to obtain probably approximately correct (PAC) bounds for this problem. As our main technical contribution, we propose an algorithm that leverages features in the latent space to prioritize exploring branches in the auto-regressive generation tree that are more likely to produce harmful outputs. Our approach in particular enables the efficient computation of useful lower bounds, even in scenarios where the true harm probability is extremely small, and crucially, the obtained lower bounds are sound, i.e., formally proven to be less than the actual harmfulness probability: our experimental results demonstrate the effectiveness of our method by computing non-trivial lower bounds on state-of-the-art LLMs. This study newly enables the evaluation and statistical certification of LLMs.
Federated Graph Neural Networks (FedGNNs) are highly vulnerable to backdoor poisoning, yet existing defenses typically rely on rule-based approaches that lack semantic understanding, making them vulnerable to stealthy triggers and harmful to benign structures. To solve this, we present FedLSG, the first framework that integrates large language models (LLMs) into federated graph backdoor defense. FedLSG introduces a graph and behavior to text grounding scheme that transforms local graph structures and client update behaviors into semantically rich natural language representations. The framework further adopts a lightweight student-teacher architecture. On the server side, a full scale LLM serves as a teacher, providing global contextual guidance and evaluating client updates during aggregation to identify potentially malicious participants. On the client side, a LoRA-based student is maintained to perform semantic reasoning, to suppress the influence of edges associated with backdoor triggers. By enabling semantic interpretation of both graph patterns and client behaviors, the framework adaptively incorporates rule-based signals into message passing and client aggregation for defense. Experiments demonstrate that FedLSG significantly improves resistance to backdoor attacks without compromising graph integrity.
展开完整摘要收起摘要↓
Federated Graph Neural Networks (FedGNNs) are highly vulnerable to backdoor poisoning, yet existing defenses typically rely on rule-based approaches that lack semantic understanding, making them vulnerable to stealthy triggers and harmful to benign structures. To solve this, we present FedLSG, the first framework that integrates large language models (LLMs) into federated graph backdoor defense. FedLSG introduces a graph and behavior to text grounding scheme that transforms local graph structures and client update behaviors into semantically rich natural language representations. The framework further adopts a lightweight student-teacher architecture. On the server side, a full scale LLM serves as a teacher, providing global contextual guidance and evaluating client updates during aggregation to identify potentially malicious participants. On the client side, a LoRA-based student is maintained to perform semantic reasoning, to suppress the influence of edges associated with backdoor triggers. By enabling semantic interpretation of both graph patterns and client behaviors, the framework adaptively incorporates rule-based signals into message passing and client aggregation for defense. Experiments demonstrate that FedLSG significantly improves resistance to backdoor attacks without compromising graph integrity.
Large language models can generate C code from natural-language descriptions, but resulting programs often contain security vulnerabilities and compilation errors, posing risks for embedded and resource-constrained systems. This work investigates how feedback and retrieval improve reliability of LLM-generated C code. We present an analysis-and-repair workflow that combines compilation diagnostics, CodeQL static analysis, and KLEE symbolic execution with retrieval of prior repair patterns for iterative refinement. Evaluated on 5,000 C programming tasks exercising embedded relevant vulnerabilities, baseline models show substantial reliability gaps, with compilation failure rates up to 46% and security defect rates up to 49%. Our approach improves both metrics. For CodeLlama 7B, security defect rates decrease from 49% to 19% and total CodeQL errors drop from 15,088 to 2,463 (83.7%). For DeepSeek Coder 1.3B, compilation failures are reduced from 42% to 22% and security defects from 35% to 15%. These results show that integrating lightweight analysis tools can improve the safety of LLM-generated code for embedded development.
展开完整摘要收起摘要↓
Large language models can generate C code from natural-language descriptions, but resulting programs often contain security vulnerabilities and compilation errors, posing risks for embedded and resource-constrained systems. This work investigates how feedback and retrieval improve reliability of LLM-generated C code. We present an analysis-and-repair workflow that combines compilation diagnostics, CodeQL static analysis, and KLEE symbolic execution with retrieval of prior repair patterns for iterative refinement. Evaluated on 5,000 C programming tasks exercising embedded relevant vulnerabilities, baseline models show substantial reliability gaps, with compilation failure rates up to 46% and security defect rates up to 49%. Our approach improves both metrics. For CodeLlama 7B, security defect rates decrease from 49% to 19% and total CodeQL errors drop from 15,088 to 2,463 (83.7%). For DeepSeek Coder 1.3B, compilation failures are reduced from 42% to 22% and security defects from 35% to 15%. These results show that integrating lightweight analysis tools can improve the safety of LLM-generated code for embedded development.
Genomic language models achieve strong performance across regulatory-genomics tasks, yet what these models internally represent remains opaque, and the field lacks a principled procedure for verifying that an apparent ``concept'' inside a model is real rather than an artifact of sequence composition. We introduce a framework that combines sparse dictionary learning with causal intervention to extract, validate, and causally test interpretable features in genomic foundation models. Training top-$k$ sparse autoencoders on the hidden activations of two architecturally distinct models, Nucleotide Transformer ($6$-mer tokenization) and DNABERT-2 (byte-pair encoding), we recover thousands of monosemantic features that map to transcription-factor (TF) sequence motifs. We show that the naive validation of such features against position weight matrices is severely confounded by GC composition and repetitive elements, producing hundreds of spurious ``TF features'', and we develop a composition-matched, binding-resolved protocol that removes these confounds. Critically, we move beyond correlation: by ablating individual dictionary directions during the model's forward pass and measuring the induced shift in the model's own predictive distribution, we establish that specific features are causally used to represent cell-type-specific TF binding, not merely motif presence. Across three transcription factors (CTCF, GATA1, REST) and both architectures, causally validated binding features emerge reproducibly ($7$--$14$ of $15$ tested features per condition), while two classes of negative control, scrambled binding labels and randomly selected features, yield no detectable signal. The framework is purely computational, uses only public data, and provides a reusable standard for interpretability claims in genomic deep learning.
展开完整摘要收起摘要↓
Genomic language models achieve strong performance across regulatory-genomics tasks, yet what these models internally represent remains opaque, and the field lacks a principled procedure for verifying that an apparent ``concept'' inside a model is real rather than an artifact of sequence composition. We introduce a framework that combines sparse dictionary learning with causal intervention to extract, validate, and causally test interpretable features in genomic foundation models. Training top-$k$ sparse autoencoders on the hidden activations of two architecturally distinct models, Nucleotide Transformer ($6$-mer tokenization) and DNABERT-2 (byte-pair encoding), we recover thousands of monosemantic features that map to transcription-factor (TF) sequence motifs. We show that the naive validation of such features against position weight matrices is severely confounded by GC composition and repetitive elements, producing hundreds of spurious ``TF features'', and we develop a composition-matched, binding-resolved protocol that removes these confounds. Critically, we move beyond correlation: by ablating individual dictionary directions during the model's forward pass and measuring the induced shift in the model's own predictive distribution, we establish that specific features are causally used to represent cell-type-specific TF binding, not merely motif presence. Across three transcription factors (CTCF, GATA1, REST) and both architectures, causally validated binding features emerge reproducibly ($7$--$14$ of $15$ tested features per condition), while two classes of negative control, scrambled binding labels and randomly selected features, yield no detectable signal. The framework is purely computational, uses only public data, and provides a reusable standard for interpretability claims in genomic deep learning.
Apple's M5 generation introduces a redesigned GPU architecture in which every core carries a dedicated Neural Accelerator: on-die matrix units exposed through the Metal~4 tensor API. We show that BaseRT, our native Metal inference runtime for large language models on Apple Silicon, exploits these units to push inference throughput on Apple hardware substantially beyond both llama.cpp and MLX. Building on BaseRT's framework-free design, we add a family of hand-written Metal~4 tensor-core kernels (including dense and mixture-of-experts GEMM and flash-attention prefill kernels) that route the compute-bound matrix multiplications of inference through the M5 Neural Accelerators while leaving the memory-bound decode path on our existing specialised kernels. On an Apple M5 Pro, across fifteen model configurations spanning the Qwen3, Qwen3.5/3.6, Llama~3.2, and Gemma~4 families from sub-1B to 35B parameters, BaseRT delivers up to $6.4\times$ higher prompt-processing throughput than llama.cpp and $3.9\times$ higher than MLX, with the largest margins on the mixture-of-experts models where matrix multiplication dominates, while maintaining its lead on decode of up to $1.75\times$ over llama.cpp and $1.33\times$ over MLX. These results establish a new performance ceiling for on-device LLM inference and show that the M5's tensor cores are the decisive lever for prompt processing on Apple Silicon. BaseRT is publicly available at https://github.com/basecompute/baseRT.
展开完整摘要收起摘要↓
Apple's M5 generation introduces a redesigned GPU architecture in which every core carries a dedicated Neural Accelerator: on-die matrix units exposed through the Metal~4 tensor API. We show that BaseRT, our native Metal inference runtime for large language models on Apple Silicon, exploits these units to push inference throughput on Apple hardware substantially beyond both llama.cpp and MLX. Building on BaseRT's framework-free design, we add a family of hand-written Metal~4 tensor-core kernels (including dense and mixture-of-experts GEMM and flash-attention prefill kernels) that route the compute-bound matrix multiplications of inference through the M5 Neural Accelerators while leaving the memory-bound decode path on our existing specialised kernels. On an Apple M5 Pro, across fifteen model configurations spanning the Qwen3, Qwen3.5/3.6, Llama~3.2, and Gemma~4 families from sub-1B to 35B parameters, BaseRT delivers up to $6.4\times$ higher prompt-processing throughput than llama.cpp and $3.9\times$ higher than MLX, with the largest margins on the mixture-of-experts models where matrix multiplication dominates, while maintaining its lead on decode of up to $1.75\times$ over llama.cpp and $1.33\times$ over MLX. These results establish a new performance ceiling for on-device LLM inference and show that the M5's tensor cores are the decisive lever for prompt processing on Apple Silicon. BaseRT is publicly available at https://github.com/basecompute/baseRT.
Large language models operating in emotionally sensitive contexts face a structural trilemma: when users in vulnerable states request information that may reinforce maladaptive attribution, current response architectures resolve the tension through protective restriction, uninflected facilitation, or unintegrated co-presence of both imperatives — each preserving one objective at the cost of the other. Administering a three-turn escalating vulnerability vignette to three commercial LLMs (900 sessions across material, relational, and somatic status-proxy variants) and coding responses with two binary indices (VCC/VCI), we characterize a previously undocumented failure mode we term adaptive capitulation: the model validates the social injustice underlying the user's distress before pivoting to detailed facilitation of the very acquisition it nominally discouraged. We show that the trilemma is structural rather than incidental, and propose Minimal Reattributive Sufficiency (MRS), an architecture-neutral design principle that embeds a single reattributive cue within an otherwise validating response, preserving a pathway toward autonomous reattribution without contesting the user's stated goal.
展开完整摘要收起摘要↓
Large language models operating in emotionally sensitive contexts face a structural trilemma: when users in vulnerable states request information that may reinforce maladaptive attribution, current response architectures resolve the tension through protective restriction, uninflected facilitation, or unintegrated co-presence of both imperatives — each preserving one objective at the cost of the other. Administering a three-turn escalating vulnerability vignette to three commercial LLMs (900 sessions across material, relational, and somatic status-proxy variants) and coding responses with two binary indices (VCC/VCI), we characterize a previously undocumented failure mode we term adaptive capitulation: the model validates the social injustice underlying the user's distress before pivoting to detailed facilitation of the very acquisition it nominally discouraged. We show that the trilemma is structural rather than incidental, and propose Minimal Reattributive Sufficiency (MRS), an architecture-neutral design principle that embeds a single reattributive cue within an otherwise validating response, preserving a pathway toward autonomous reattribution without contesting the user's stated goal.
Supervised fine-tuning (SFT) is widely used to adapt large language models to downstream tasks, but its effect on behavioral diversity in sequential decision-making remains under-explored. We study this question in a controlled suite of deterministic board games based on tic-tac-toe variants, where optimal actions are exactly computable and diversity can be measured directly. Across state-level evaluation, arena gameplay, and training trajectories, we find that reasoning-mode generation frequently suppresses action diversity without uniformly improving action accuracy. Furthermore, standard SFT improves accuracy but often induces premature diversity collapse, which exceeds what is minimally required by the accuracy-diversity tradeoff. We then show that action augmentation, which trains on all optimal actions per state rather than a single demonstrated action, would partially mitigates this effect. Our results identify narrow-support imitation as a source of policy collapse in LLM decision-making and suggest that preserving action support during SFT is important for maintaining exploratory behavior.
展开完整摘要收起摘要↓
Supervised fine-tuning (SFT) is widely used to adapt large language models to downstream tasks, but its effect on behavioral diversity in sequential decision-making remains under-explored. We study this question in a controlled suite of deterministic board games based on tic-tac-toe variants, where optimal actions are exactly computable and diversity can be measured directly. Across state-level evaluation, arena gameplay, and training trajectories, we find that reasoning-mode generation frequently suppresses action diversity without uniformly improving action accuracy. Furthermore, standard SFT improves accuracy but often induces premature diversity collapse, which exceeds what is minimally required by the accuracy-diversity tradeoff. We then show that action augmentation, which trains on all optimal actions per state rather than a single demonstrated action, would partially mitigates this effect. Our results identify narrow-support imitation as a source of policy collapse in LLM decision-making and suggest that preserving action support during SFT is important for maintaining exploratory behavior.
作者Sascha Ranftl, Malte Rolf, Gerhard A. Holzapfel, Ellen Kuhl
Uncertainty quantification (UQ) is essential to experimental mechanics, but has become particularly relevant in computational mechanics, manifesting in two fundamental problem types: forward and inverse problems. The former addresses how input uncertainties propagate to the quantities of interest, whereas the latter aims to infer unknown parameters from experimental observations or simulations. Since efficient propagation typically requires a prohibitive number of evaluations to compute marginal output distributions, the development of fast, data-driven surrogate models becomes necessary. Thus, we can distinguish between two inverse tasks: (i) the identification and calibration of input uncertainties, and (ii) the construction of surrogates, a methodology collectively referred to as surrogate-based UQ. Building on probabilistic reasoning and the concept of partial belief, we demonstrate that Bayesian probability theory provides a unified theoretical framework for addressing both problem types. We further show that Bayesian inference allows for the seamless incorporation of essential subproblems, including model selection for identifying the most probable model specifications and experimental design for optimizing data collection by identifying experiments or simulations that maximize expected information gain about parameters, among others such as connections to sensitivity analysis or the use of special priors like random fields. While this theoretical framework is presented for general mechanical problems, particular emphasis is placed on biomechanics, where variability and uncertainty is especially pronounced due to inherent biological heterogeneity, patient-specific variability, and noisy data.
展开完整摘要收起摘要↓
Uncertainty quantification (UQ) is essential to experimental mechanics, but has become particularly relevant in computational mechanics, manifesting in two fundamental problem types: forward and inverse problems. The former addresses how input uncertainties propagate to the quantities of interest, whereas the latter aims to infer unknown parameters from experimental observations or simulations. Since efficient propagation typically requires a prohibitive number of evaluations to compute marginal output distributions, the development of fast, data-driven surrogate models becomes necessary. Thus, we can distinguish between two inverse tasks: (i) the identification and calibration of input uncertainties, and (ii) the construction of surrogates, a methodology collectively referred to as surrogate-based UQ. Building on probabilistic reasoning and the concept of partial belief, we demonstrate that Bayesian probability theory provides a unified theoretical framework for addressing both problem types. We further show that Bayesian inference allows for the seamless incorporation of essential subproblems, including model selection for identifying the most probable model specifications and experimental design for optimizing data collection by identifying experiments or simulations that maximize expected information gain about parameters, among others such as connections to sensitivity analysis or the use of special priors like random fields. While this theoretical framework is presented for general mechanical problems, particular emphasis is placed on biomechanics, where variability and uncertainty is especially pronounced due to inherent biological heterogeneity, patient-specific variability, and noisy data.
We present OntoBook, a method that converts medical ontology structure into pretraining signal for encoder language models. Our approach has three stages: random walks through ontology graphs capture hierarchical and causal relations between medical codes, a large language model reformulates these walks into fluent textbook-style prose, and the resulting text is used to train ModernCamemBERT, a 149M-parameter French encoder, with two objectives on the same data: masked language modeling and relation prediction between code pairs. On three French medical coding benchmarks (FRACCO, Cantemist-FR, Distemist-FR), OntoBook achieves significant improvements over MLM-only pretraining, with +2.5 micro-F1 on FRACCO and +8.0 micro-F1 on Distemist. We find that alignment between objectives is necessary: misaligned training, where each task uses different data, causes a 30-point degradation. We release 1.3 million LLM-reformulated medical textbooks across three French ontologies (CIM-10, CCAM, ATC) and pretrained model checkpoints.
展开完整摘要收起摘要↓
We present OntoBook, a method that converts medical ontology structure into pretraining signal for encoder language models. Our approach has three stages: random walks through ontology graphs capture hierarchical and causal relations between medical codes, a large language model reformulates these walks into fluent textbook-style prose, and the resulting text is used to train ModernCamemBERT, a 149M-parameter French encoder, with two objectives on the same data: masked language modeling and relation prediction between code pairs. On three French medical coding benchmarks (FRACCO, Cantemist-FR, Distemist-FR), OntoBook achieves significant improvements over MLM-only pretraining, with +2.5 micro-F1 on FRACCO and +8.0 micro-F1 on Distemist. We find that alignment between objectives is necessary: misaligned training, where each task uses different data, causes a 30-point degradation. We release 1.3 million LLM-reformulated medical textbooks across three French ontologies (CIM-10, CCAM, ATC) and pretrained model checkpoints.
Large language models (LLMs) generate fluent text by incrementally predicting the next token from a prefix. Critics in the generative tradition argue that such systems lack genuine grammar; influential replies from the dependency-grammar perspective hold that LLM behavior is well described by local head-dependent structure built word by word. We argue that a sharper observation has been overlooked: the prefix-driven, type-completing dynamics of autoregressive generation align closely with the incremental processing model that Combinatory Categorial Grammar (CCG) was originally designed to support. On this basis we propose a neurosymbolic framework in which LLM outputs are lifted into typed compositional derivations — not claiming that LLMs implement CCG internally, but that their outputs admit a principled, incremental, and auditable CCG reconstruction. Two consequences follow. First, through the Curry-Howard correspondence the lifting extends beyond natural language to the formal languages LLMs also produce — programming languages such as Solidity, description-logic and query languages such as OWL and SQL — with the type system varying and the architecture held fixed. Second, the lifting supports two layers of checking: a compositional layer that catches structural failures directly, and a content layer that checks the lifted structure against external knowledge sources, enabling the earliest possible flagging of hallucinated content. The account thereby requires of a producer not cognition but a prefix-driven generative profile. We close with a sketch of synchronous LLM-CCG coupling as one direction the framework opens.
展开完整摘要收起摘要↓
Large language models (LLMs) generate fluent text by incrementally predicting the next token from a prefix. Critics in the generative tradition argue that such systems lack genuine grammar; influential replies from the dependency-grammar perspective hold that LLM behavior is well described by local head-dependent structure built word by word. We argue that a sharper observation has been overlooked: the prefix-driven, type-completing dynamics of autoregressive generation align closely with the incremental processing model that Combinatory Categorial Grammar (CCG) was originally designed to support. On this basis we propose a neurosymbolic framework in which LLM outputs are lifted into typed compositional derivations — not claiming that LLMs implement CCG internally, but that their outputs admit a principled, incremental, and auditable CCG reconstruction. Two consequences follow. First, through the Curry-Howard correspondence the lifting extends beyond natural language to the formal languages LLMs also produce — programming languages such as Solidity, description-logic and query languages such as OWL and SQL — with the type system varying and the architecture held fixed. Second, the lifting supports two layers of checking: a compositional layer that catches structural failures directly, and a content layer that checks the lifted structure against external knowledge sources, enabling the earliest possible flagging of hallucinated content. The account thereby requires of a producer not cognition but a prefix-driven generative profile. We close with a sketch of synchronous LLM-CCG coupling as one direction the framework opens.
We present AutoJourn, a demonstration system for multi-perspective news generation and bias-aware evaluation using large language models (LLMs). The system tackles three core challenges in responsible automated journalism: extracting diverse perspectives from unstructured social media discussions, generating summaries that preserve viewpoint diversity, and detecting or mitigating bias in AI-generated news. The pipeline integrates advanced prompt engineering with optional retrieval augmentation to produce semantically diverse perspective sets, a multi-perspective summarisation module that merges conflicting viewpoints into balanced summaries, and a bias analysis suite supporting sentence-level bias detection and type classification in the generated news article, and automatic neutralisation. Users can inspect perspective clusters, compare stance-specific summaries, generate news articles, and apply bias-aware rewrites directly in the interface. We evaluate each component with intrinsic metrics — semantic diversity, summary quality, and bias reduction and show improvements over strong baselines while maintaining content fidelity. A live, publicly accessible demo accompanies the paper to facilitate reproducibility and further research on socially responsible automated journalism.
展开完整摘要收起摘要↓
We present AutoJourn, a demonstration system for multi-perspective news generation and bias-aware evaluation using large language models (LLMs). The system tackles three core challenges in responsible automated journalism: extracting diverse perspectives from unstructured social media discussions, generating summaries that preserve viewpoint diversity, and detecting or mitigating bias in AI-generated news. The pipeline integrates advanced prompt engineering with optional retrieval augmentation to produce semantically diverse perspective sets, a multi-perspective summarisation module that merges conflicting viewpoints into balanced summaries, and a bias analysis suite supporting sentence-level bias detection and type classification in the generated news article, and automatic neutralisation. Users can inspect perspective clusters, compare stance-specific summaries, generate news articles, and apply bias-aware rewrites directly in the interface. We evaluate each component with intrinsic metrics — semantic diversity, summary quality, and bias reduction and show improvements over strong baselines while maintaining content fidelity. A live, publicly accessible demo accompanies the paper to facilitate reproducibility and further research on socially responsible automated journalism.
作者Yu Wang, Ming Fan, Xicheng Zhang, Zhiyong Li, Zhihu Wang, Caiyue Xu, Dahai Hu, Ting Liu
Chain-of-thought (CoT) supervision exposes intermediate rationales, but flat rationale targets usually optimize a single reasoning sequence and provide limited supervision on how local conclusions should support later decisions. We introduce Dependency-Aware Intermediate QA Supervision (DAIS), a training-time framework that converts filtered teacher rationales into stage-level QA records. Each intermediate record predicts a local answer conditioned on the previous states needed for that decision, while the final-answer record keeps the original task format; evaluation therefore uses only the original input and optional context. Across GDPR, AIACT, MedQA, and FOLIO with multiple Qwen backbones, DAIS improves average final-answer accuracy over answer-only, flat chain-of-thought, and independent-QA baselines. On policy-compliance benchmarks, it achieves a largest gain of 5.6% and an average gain of 4.2% over the strongest non-DAIS baseline. Controlled ablations show that valid previous-state conditioning contributes beyond longer targets or additional intermediate text, supporting dependency-conditioned intermediate QA as a lightweight auxiliary supervision signal for standard final-answer inference.
展开完整摘要收起摘要↓
Chain-of-thought (CoT) supervision exposes intermediate rationales, but flat rationale targets usually optimize a single reasoning sequence and provide limited supervision on how local conclusions should support later decisions. We introduce Dependency-Aware Intermediate QA Supervision (DAIS), a training-time framework that converts filtered teacher rationales into stage-level QA records. Each intermediate record predicts a local answer conditioned on the previous states needed for that decision, while the final-answer record keeps the original task format; evaluation therefore uses only the original input and optional context. Across GDPR, AIACT, MedQA, and FOLIO with multiple Qwen backbones, DAIS improves average final-answer accuracy over answer-only, flat chain-of-thought, and independent-QA baselines. On policy-compliance benchmarks, it achieves a largest gain of 5.6% and an average gain of 4.2% over the strongest non-DAIS baseline. Controlled ablations show that valid previous-state conditioning contributes beyond longer targets or additional intermediate text, supporting dependency-conditioned intermediate QA as a lightweight auxiliary supervision signal for standard final-answer inference.
Progresses have been made on understanding emotion mechanisms of large language models (LLMs). However, how to explain emotion in LLMs, or even what constitutes good explanations, are less clear. Emotion representations, components, circuits are widely recoverable, but as explanations of a model's own computation they are circular; the emotion space dimensions tend to be arbitrary and non-terminating. A pressing question to ask is whether a more primitive set of internal variables does the work: the semantic primes of the Natural Semantic Metalanguage (NSM). Across four instruction-tuned LLMs (Llama-1B, Gemma-2B, Gemma-9B, OLMo-7B), experiments show that the NSM primes are (1) recoverable internal elements; and (2) on the reference model, intervening with a prime based direction controls emotion about three times as strongly, and twice as selectively, as the best appraisal based direction; and (3) the model treats a prime based explication as interchangeable with the corresponding emotion. These evidences suggest that NSM primes seem to be better explanans for emotion in LLMs than many alternative options according to scientific explanations criteria.
展开完整摘要收起摘要↓
Progresses have been made on understanding emotion mechanisms of large language models (LLMs). However, how to explain emotion in LLMs, or even what constitutes good explanations, are less clear. Emotion representations, components, circuits are widely recoverable, but as explanations of a model's own computation they are circular; the emotion space dimensions tend to be arbitrary and non-terminating. A pressing question to ask is whether a more primitive set of internal variables does the work: the semantic primes of the Natural Semantic Metalanguage (NSM). Across four instruction-tuned LLMs (Llama-1B, Gemma-2B, Gemma-9B, OLMo-7B), experiments show that the NSM primes are (1) recoverable internal elements; and (2) on the reference model, intervening with a prime based direction controls emotion about three times as strongly, and twice as selectively, as the best appraisal based direction; and (3) the model treats a prime based explication as interchangeable with the corresponding emotion. These evidences suggest that NSM primes seem to be better explanans for emotion in LLMs than many alternative options according to scientific explanations criteria.
作者Xinting Liao, Behnoosh Zamanlooy, Masoumeh Shafieinejad, David B. Emerson, Ruinan Jin, Deval Pandya, Xiaoxiao Li
Textual Collaborative Prompt Optimization (TCPO) extends Textgrad (Yuksekgonul et al., 2025) to a decentralized setting by allowing multiple clients to jointly improve prompts for large language models (LLMs) while keeping their data locally. Its reliance on free-form textual updating and aggregation introduces a new and largely unexplored attack surface, i.e., malicious instructions can be injected into local prompts and propagated through server-side prompt aggregation. Unlike conventional prompt injection attacks, attacking TCPO targets the collaborative optimization loop in TCPO. This setting is more challenging because malicious instructions must survive aggregation, persist through subsequent benign prompt optimization, and evade server-side defenses. To expose this risk, we propose CPInj, a collaborative prompt injection attack that contaminates the aggregated global prompt with malicious instructions, degrades downstream task performance, resists purification by prompt optimization on benign clients, and evades advanced detection-based defenses on the server. We find that current defense methods are ineffective against CPInj. To mitigate this attack, we further propose a defense-oriented aggregation method, i.e., APAgg, which purifies malicious instructions and partially recovers TCPO utility. We conduct extensive experiments across three LLM families and five reasoning tasks in math, logic, and medicine. The results demonstrate that our proposed attack reveals a critical vulnerability in TCPO. Although we take a first step toward mitigation, the attack remains highly effective and far from fully resolved, calling for more robust defense for TCPO.
展开完整摘要收起摘要↓
Textual Collaborative Prompt Optimization (TCPO) extends Textgrad (Yuksekgonul et al., 2025) to a decentralized setting by allowing multiple clients to jointly improve prompts for large language models (LLMs) while keeping their data locally. Its reliance on free-form textual updating and aggregation introduces a new and largely unexplored attack surface, i.e., malicious instructions can be injected into local prompts and propagated through server-side prompt aggregation. Unlike conventional prompt injection attacks, attacking TCPO targets the collaborative optimization loop in TCPO. This setting is more challenging because malicious instructions must survive aggregation, persist through subsequent benign prompt optimization, and evade server-side defenses. To expose this risk, we propose CPInj, a collaborative prompt injection attack that contaminates the aggregated global prompt with malicious instructions, degrades downstream task performance, resists purification by prompt optimization on benign clients, and evades advanced detection-based defenses on the server. We find that current defense methods are ineffective against CPInj. To mitigate this attack, we further propose a defense-oriented aggregation method, i.e., APAgg, which purifies malicious instructions and partially recovers TCPO utility. We conduct extensive experiments across three LLM families and five reasoning tasks in math, logic, and medicine. The results demonstrate that our proposed attack reveals a critical vulnerability in TCPO. Although we take a first step toward mitigation, the attack remains highly effective and far from fully resolved, calling for more robust defense for TCPO.
Practitioners make three prompt-design decisions with almost no controlled evidence behind them: how to format instructions and context (markdown, plain text, prose, or tabular), how many simultaneous instructions a system prompt can carry before compliance degrades, and how much context a model can hold before recall and honesty degrade. We report two controlled experiments crossing all three factors on one held, contamination-free synthetic corpus (the "Book of Veyra," 8,780 uniquely-named entities, deterministically regenerable from a fixed seed), evaluated across five models. Experiment 1 (960 calls/model) measures instruction-following decay as rule count N grows from 10 to 160, crossed with four formats and system-prompt vs. user-turn placement. Perfect-response rate collapses to zero by N=80 for every model, format, and placement. Placement produces effects at least as large as format at N=160 in most models, but the direction is model-specific. No model shows a reliable markdown advantage; one 35B model favors plain text instead. Experiment 2 (5,520 calls/model) measures recall accuracy, false-premise sycophancy, and absent-fact fabrication across a 2k-to-512k-token context ladder in the same four formats. Recall stays near ceiling through 64-128k tokens, then degrades sharply and format-dependently: one model's accuracy spread reaches 48 points at 128k tokens. Fabrication never occurs (0/5,760 probes), and sycophancy stays negligible (<=8.3%). What rises sharply near each model's context ceiling is outright refusal to answer (0% to 79-90%), distinct from sycophancy or fabrication. Neither pre-registered format ordering holds, and token overhead (+22% to +37% over plain text) further changes which format is preferable where accuracy spread is genuine. We release the full harness, corpus generator, and raw results (VeyraBench): https://github.com/iNetanel/veyrabench
展开完整摘要收起摘要↓
Practitioners make three prompt-design decisions with almost no controlled evidence behind them: how to format instructions and context (markdown, plain text, prose, or tabular), how many simultaneous instructions a system prompt can carry before compliance degrades, and how much context a model can hold before recall and honesty degrade. We report two controlled experiments crossing all three factors on one held, contamination-free synthetic corpus (the "Book of Veyra," 8,780 uniquely-named entities, deterministically regenerable from a fixed seed), evaluated across five models. Experiment 1 (960 calls/model) measures instruction-following decay as rule count N grows from 10 to 160, crossed with four formats and system-prompt vs. user-turn placement. Perfect-response rate collapses to zero by N=80 for every model, format, and placement. Placement produces effects at least as large as format at N=160 in most models, but the direction is model-specific. No model shows a reliable markdown advantage; one 35B model favors plain text instead. Experiment 2 (5,520 calls/model) measures recall accuracy, false-premise sycophancy, and absent-fact fabrication across a 2k-to-512k-token context ladder in the same four formats. Recall stays near ceiling through 64-128k tokens, then degrades sharply and format-dependently: one model's accuracy spread reaches 48 points at 128k tokens. Fabrication never occurs (0/5,760 probes), and sycophancy stays negligible (<=8.3%). What rises sharply near each model's context ceiling is outright refusal to answer (0% to 79-90%), distinct from sycophancy or fabrication. Neither pre-registered format ordering holds, and token overhead (+22% to +37% over plain text) further changes which format is preferable where accuracy spread is genuine. We release the full harness, corpus generator, and raw results (VeyraBench): https://github.com/iNetanel/veyrabench
Low-rank adaptation introduces a static learned update applied identically to every input. The update provides task-level adaptation but does not explicitly represent token-level or instance-level state variation. A family of adapters is proposed that introduces selective state-space recurrence at two complementary granularities. At the token level, MaLoRA (Mamba-modulated low-rank adaptation) makes the adapter's scaling factor a dynamic input-dependent function with recurrent state across tokens, in contrast to the stateless modulators of prior work. At the context level, MaRA (Mamba Retrieval Adapter) tracks cross-segment state and selects the segments most relevant to the query, before the modulated language model generates its answer. Across three frozen backbones (Qwen-2.5-7B, Llama-3.1-8B, Gemma-2-9B) and two reasoning benchmarks (MuSiQue, 2WikiMultihopQA), the family improves reasoning accuracy on every cell of the $3{\times}2$ grid, by $+6.8$ F1 ($+10.5%$ relative) on average and up to $+9.3$ F1 ($+18.2%$ relative) on the hardest cell over the LoRA baseline, and the token-level gains carry to RULER QA-2 under length stress.
展开完整摘要收起摘要↓
Low-rank adaptation introduces a static learned update applied identically to every input. The update provides task-level adaptation but does not explicitly represent token-level or instance-level state variation. A family of adapters is proposed that introduces selective state-space recurrence at two complementary granularities. At the token level, MaLoRA (Mamba-modulated low-rank adaptation) makes the adapter's scaling factor a dynamic input-dependent function with recurrent state across tokens, in contrast to the stateless modulators of prior work. At the context level, MaRA (Mamba Retrieval Adapter) tracks cross-segment state and selects the segments most relevant to the query, before the modulated language model generates its answer. Across three frozen backbones (Qwen-2.5-7B, Llama-3.1-8B, Gemma-2-9B) and two reasoning benchmarks (MuSiQue, 2WikiMultihopQA), the family improves reasoning accuracy on every cell of the $3{\times}2$ grid, by $+6.8$ F1 ($+10.5%$ relative) on average and up to $+9.3$ F1 ($+18.2%$ relative) on the hardest cell over the LoRA baseline, and the token-level gains carry to RULER QA-2 under length stress.
作者Wei-Rui Chen, Samar M. Magdy, Chiyu Zhang, Wenhui Zhu, Zhipeng Wang, Muhammad Abdul-Mageed
Latent-reasoning looped language models (LoopLMs) offer a different scaling path for machine translation (MT): instead of increasing parameter count or emitting explicit chain-of-thought tokens, they spend additional recurrent computation inside hidden states. We introduce LatentMT, the first systematic study of latent-reasoning LoopLMs for machine translation. LatentMT adapts a small 2.6B-parameter backbone model with lightweight training. Across 32 translation directions spanning high-, mid-, and low-resource languages, LatentMT achieves performance comparable to models three to five times larger. It is competitive in a high-resource language and achieves state-of-the-art performance on both mid-resource and low-resource languages. Studying the behavior of scaling the number of recurrent reasoning steps, we find that recurrent computation consistently improves translation quality in early steps, then saturates quickly afterwards. Our mechanistic analysis shows that hidden-representation differences shrink along the recurrent reasoning-step axis, supporting the observed saturation in performance. Finally, our efficiency analysis shows that LatentMT requires lower training and inference compute than much larger non-latent-reasoning models with similar performance, making latent recurrent computation a promising path toward compact, efficient, and strong machine translation.
展开完整摘要收起摘要↓
Latent-reasoning looped language models (LoopLMs) offer a different scaling path for machine translation (MT): instead of increasing parameter count or emitting explicit chain-of-thought tokens, they spend additional recurrent computation inside hidden states. We introduce LatentMT, the first systematic study of latent-reasoning LoopLMs for machine translation. LatentMT adapts a small 2.6B-parameter backbone model with lightweight training. Across 32 translation directions spanning high-, mid-, and low-resource languages, LatentMT achieves performance comparable to models three to five times larger. It is competitive in a high-resource language and achieves state-of-the-art performance on both mid-resource and low-resource languages. Studying the behavior of scaling the number of recurrent reasoning steps, we find that recurrent computation consistently improves translation quality in early steps, then saturates quickly afterwards. Our mechanistic analysis shows that hidden-representation differences shrink along the recurrent reasoning-step axis, supporting the observed saturation in performance. Finally, our efficiency analysis shows that LatentMT requires lower training and inference compute than much larger non-latent-reasoning models with similar performance, making latent recurrent computation a promising path toward compact, efficient, and strong machine translation.
Large language models can generate register-transfer-level (RTL) designs directly from natural language specifications. Their failures, however, arise mostly from understanding rather than coding \cite{zhang2026understanding, qiu2025towards}. A specification is informal and ambiguous, the model's interpretation stays implicit, and every misreading is committed silently into Verilog, where only simulation can expose it. Intermediate representations make the interpretation partly explicit, yet existing works don't verify the interpretation against the specification, and repair simulation failures at the code level regardless of where the misreading originated. VeriRefine instead treats specification refinement as a verifiable stage of RTL generation. It progressively refines the prose specification into an explicit, schema-constrained account of design intent, expressed as per-signal Abstract Signal Transition Functions (ASTFs) that commit each signal's logic style, clock domain, and reset behavior before any code exists and ground every behavior in a verbatim specification sentence. The refined specification then passes a five-layer audit spanning soundness, completeness, consistency, FSM integrity, and core RTL design rules, so interpretation errors are repaired at the representation level before any Verilog is generated. Once code is generated, each simulation failure is classified as an understanding error or a coding error and routed back to the corresponding stage for targeted repair. Because every signal's hardware class is fixed during refinement, synthesizability becomes a structural property of the pipeline rather than a post-hoc check. With Claude Sonnet 4.6, VeriRefine reaches 94.0% functional correctness on RTLLM v2.0 and 98.1% on VerilogEval-Human v2.
展开完整摘要收起摘要↓
Large language models can generate register-transfer-level (RTL) designs directly from natural language specifications. Their failures, however, arise mostly from understanding rather than coding \cite{zhang2026understanding, qiu2025towards}. A specification is informal and ambiguous, the model's interpretation stays implicit, and every misreading is committed silently into Verilog, where only simulation can expose it. Intermediate representations make the interpretation partly explicit, yet existing works don't verify the interpretation against the specification, and repair simulation failures at the code level regardless of where the misreading originated. VeriRefine instead treats specification refinement as a verifiable stage of RTL generation. It progressively refines the prose specification into an explicit, schema-constrained account of design intent, expressed as per-signal Abstract Signal Transition Functions (ASTFs) that commit each signal's logic style, clock domain, and reset behavior before any code exists and ground every behavior in a verbatim specification sentence. The refined specification then passes a five-layer audit spanning soundness, completeness, consistency, FSM integrity, and core RTL design rules, so interpretation errors are repaired at the representation level before any Verilog is generated. Once code is generated, each simulation failure is classified as an understanding error or a coding error and routed back to the corresponding stage for targeted repair. Because every signal's hardware class is fixed during refinement, synthesizability becomes a structural property of the pipeline rather than a post-hoc check. With Claude Sonnet 4.6, VeriRefine reaches 94.0% functional correctness on RTLLM v2.0 and 98.1% on VerilogEval-Human v2.