Advanced chart question answering requires both precise perception of small visual elements and multi-step reasoning across several subplots. While existing MLLMs are strong at understanding single plots, they often struggle with multi-step reasoning across multiple subplots. We propose HierVA, a hierarchical visual agent framework for chart reasoning that iteratively constructs and updates a working context in a joint image--text space. A high-level manager generates plans and maintains a compact context containing only key information, while specialized workers perform reasoning, gather evidence, and return results. In particular, the agent maintains separate visual and textual contexts, using a zoom-in tool to restrict the visual context. Experiments on the CharXiv reasoning subset demonstrate consistent improvements over strong multimodal baselines, and ablation studies verify that hierarchical architecture, scoped visual context, and distilled context contribute complementary gains.
展开完整摘要收起摘要↓
Advanced chart question answering requires both precise perception of small visual elements and multi-step reasoning across several subplots. While existing MLLMs are strong at understanding single plots, they often struggle with multi-step reasoning across multiple subplots. We propose HierVA, a hierarchical visual agent framework for chart reasoning that iteratively constructs and updates a working context in a joint image--text space. A high-level manager generates plans and maintains a compact context containing only key information, while specialized workers perform reasoning, gather evidence, and return results. In particular, the agent maintains separate visual and textual contexts, using a zoom-in tool to restrict the visual context. Experiments on the CharXiv reasoning subset demonstrate consistent improvements over strong multimodal baselines, and ablation studies verify that hierarchical architecture, scoped visual context, and distilled context contribute complementary gains.
作者Yixuan Mei, Zikun Li, Zixuan Chen, Shiqi Pan, Mengdi Wu, Xupeng Miao, Zhihao Jia, K. V. Rashmi
The usage of large language models (LLMs) has grown increasingly fragmented, with no single model dominating. Meanwhile, cloud providers offer a wide range of mid-tier and older-generation GPUs that enjoy better availability and deliver comparable performance per dollar to top-tier hardware. To efficiently harness these heterogeneous resources for serving multiple LLMs concurrently, we introduce Coral, an adaptive heterogeneity-aware multi-LLM serving system. The key idea behind Coral is to jointly optimize resource allocation and the serving strategy of each model replica across all models. To keep pace with shifting throughput demand and resource availability, Coral applies a lossless two-stage decomposition that preserves joint optimality while cutting online solve time from hours to tens of seconds. Our evaluation across 6 models and 20 GPU configurations shows that Coral reduces serving cost by up to 2.79$\times$ over the best baseline, and delivers up to 2.39$\times$ higher goodput under scarce resource availability.
展开完整摘要收起摘要↓
The usage of large language models (LLMs) has grown increasingly fragmented, with no single model dominating. Meanwhile, cloud providers offer a wide range of mid-tier and older-generation GPUs that enjoy better availability and deliver comparable performance per dollar to top-tier hardware. To efficiently harness these heterogeneous resources for serving multiple LLMs concurrently, we introduce Coral, an adaptive heterogeneity-aware multi-LLM serving system. The key idea behind Coral is to jointly optimize resource allocation and the serving strategy of each model replica across all models. To keep pace with shifting throughput demand and resource availability, Coral applies a lossless two-stage decomposition that preserves joint optimality while cutting online solve time from hours to tens of seconds. Our evaluation across 6 models and 20 GPU configurations shows that Coral reduces serving cost by up to 2.79$\times$ over the best baseline, and delivers up to 2.39$\times$ higher goodput under scarce resource availability.
Evaluating generative video models remains an open problem. Reference-based metrics such as Structural Similarity Index Measure (SSIM) and Peak Signal to Noise Ratio (PSNR) reward pixel fidelity over semantic correctness, while Frechet Video Distance (FVD) favors distributional textures over physical plausibility. Binary Visual Question Answering (VQA) based benchmarks like VBench~2.0 are prone to yes-bias and rely on low-resolution auditors that miss temporal failures. Moreover, their prompts target a single dimension at a time, multiplying the number of videos required while still not guaranteeing reliable results. WorldJen addresses these limitations directly. Binary VQA is replaced with Likert-scale questionnaires graded by a VLM that receives frames at native video resolution. Video generation costs are addressed by using adversarially curated prompts that are designed to exercise up to 16 quality dimensions simultaneously. The framework is built around two interlocking contributions. First, A blind human preference study is conducted, accumulating (2,696 pairwise annotations from 7 annotators with 100% pair coverage over 50 of the curated prompts $\times$ 6 state-of-the-art video models. A mean inter-annotator agreement of 66.9% is achieved and the study establishes a human ground-truth Bradley-Terry (BT) rating with a three-tier structure. Second, A VLM-as-a-judge evaluation engine using prompt-specific, dimension-specific Likert questionnaires (10 questions per dimension, 47,160 scored responses) judges the videos and reproduces the human-established three-tier BT rating structure independently. The VLM achieves a Spearman $\hatρ=1.000,~p=0.0014$ that is interpreted as tier agreement with the human results. Six focused ablation studies validate the robustness of the VLM evaluation framework. Project page: https://moonmath.ai/worldjen/
展开完整摘要收起摘要↓
Evaluating generative video models remains an open problem. Reference-based metrics such as Structural Similarity Index Measure (SSIM) and Peak Signal to Noise Ratio (PSNR) reward pixel fidelity over semantic correctness, while Frechet Video Distance (FVD) favors distributional textures over physical plausibility. Binary Visual Question Answering (VQA) based benchmarks like VBench~2.0 are prone to yes-bias and rely on low-resolution auditors that miss temporal failures. Moreover, their prompts target a single dimension at a time, multiplying the number of videos required while still not guaranteeing reliable results. WorldJen addresses these limitations directly. Binary VQA is replaced with Likert-scale questionnaires graded by a VLM that receives frames at native video resolution. Video generation costs are addressed by using adversarially curated prompts that are designed to exercise up to 16 quality dimensions simultaneously. The framework is built around two interlocking contributions. First, A blind human preference study is conducted, accumulating (2,696 pairwise annotations from 7 annotators with 100% pair coverage over 50 of the curated prompts $\times$ 6 state-of-the-art video models. A mean inter-annotator agreement of 66.9% is achieved and the study establishes a human ground-truth Bradley-Terry (BT) rating with a three-tier structure. Second, A VLM-as-a-judge evaluation engine using prompt-specific, dimension-specific Likert questionnaires (10 questions per dimension, 47,160 scored responses) judges the videos and reproduces the human-established three-tier BT rating structure independently. The VLM achieves a Spearman $\hatρ=1.000,~p=0.0014$ that is interpreted as tier agreement with the human results. Six focused ablation studies validate the robustness of the VLM evaluation framework. Project page: https://moonmath.ai/worldjen/
Achieving Artificial General Intelligence (AGI) requires agents that learn and interact adaptively, with interactive world models providing scalable environments for perception, reasoning, and action. Yet current research still lacks large-scale datasets and unified benchmarks to evaluate their physical interaction capabilities. To address this, we propose iWorld-Bench, a comprehensive benchmark for training and testing world models on interaction-related abilities such as distance perception and memory. We construct a diverse dataset with 330k video clips and select 2.1k high-quality samples covering varied perspectives, weather, and scenes. As existing world models differ in interaction modalities, we introduce an Action Generation Framework to unify evaluation and design six task types, generating 4.9k test samples. These tasks jointly assess model performance across visual generation, trajectory following, and memory. Evaluating 14 representative world models, we identify key limitations and provide insights for future research. The iWorld-Bench model leaderboard is publicly available at iWorld-Bench.com.
展开完整摘要收起摘要↓
Achieving Artificial General Intelligence (AGI) requires agents that learn and interact adaptively, with interactive world models providing scalable environments for perception, reasoning, and action. Yet current research still lacks large-scale datasets and unified benchmarks to evaluate their physical interaction capabilities. To address this, we propose iWorld-Bench, a comprehensive benchmark for training and testing world models on interaction-related abilities such as distance perception and memory. We construct a diverse dataset with 330k video clips and select 2.1k high-quality samples covering varied perspectives, weather, and scenes. As existing world models differ in interaction modalities, we introduce an Action Generation Framework to unify evaluation and design six task types, generating 4.9k test samples. These tasks jointly assess model performance across visual generation, trajectory following, and memory. Evaluating 14 representative world models, we identify key limitations and provide insights for future research. The iWorld-Bench model leaderboard is publicly available at iWorld-Bench.com.
Reinforcement learning with verifiable rewards has been used to elicit impressive performance from language models in many domains. But, broadly beneficial deployments of AI may require us to train models with strong capabilities in "fuzzy", hard-to-supervise domains. In this paper, we develop methods to align language models in fuzzy domains where human experts are still able to provide high-quality supervision signal, but only for a small number of model outputs, using online natural language feedback. Specifically, we train models by iteratively optimizing against proxy reward signals, stopping at the point of over-optimization, collecting fresh expert supervision, and updating the proxy reward. We construct proxy reward models from language models using in-context learning (ICL) and fine-tuning. We test our methods by eliciting creative writing and alignment research capabilities in Qwen3-8B and Haiku 4.5 respectively. For Qwen3-8B, ICL methods recover up to 35% of performance with 50x fewer expert samples, while fine-tuning methods recover 80% with up to 20x fewer samples and 100% with 3x fewer samples. For Haiku 4.5, ICL methods recover up to 35% of performance with 30x fewer samples, and fine-tuning methods recover 100% with 10x fewer samples. Our results suggest that online natural language feedback can substantially improve the data efficiency of expert supervision.
展开完整摘要收起摘要↓
Reinforcement learning with verifiable rewards has been used to elicit impressive performance from language models in many domains. But, broadly beneficial deployments of AI may require us to train models with strong capabilities in "fuzzy", hard-to-supervise domains. In this paper, we develop methods to align language models in fuzzy domains where human experts are still able to provide high-quality supervision signal, but only for a small number of model outputs, using online natural language feedback. Specifically, we train models by iteratively optimizing against proxy reward signals, stopping at the point of over-optimization, collecting fresh expert supervision, and updating the proxy reward. We construct proxy reward models from language models using in-context learning (ICL) and fine-tuning. We test our methods by eliciting creative writing and alignment research capabilities in Qwen3-8B and Haiku 4.5 respectively. For Qwen3-8B, ICL methods recover up to 35% of performance with 50x fewer expert samples, while fine-tuning methods recover 80% with up to 20x fewer samples and 100% with 3x fewer samples. For Haiku 4.5, ICL methods recover up to 35% of performance with 30x fewer samples, and fine-tuning methods recover 100% with 10x fewer samples. Our results suggest that online natural language feedback can substantially improve the data efficiency of expert supervision.
We introduce a simple yet powerful framework for training large language models. In contrast to the standard autoregressive next-token prediction based on an exact prefix, we propose a perturbation-based procedure that first transforms the prefix into a semantic neighbor and then conditions on this perturbed variant for next-token prediction. This yields a hierarchical model with a pre-post-additive noise structure. Within this framework, we develop a rigorous theory of extrapolability, namely, the capacity of a model class to make reliable predictions for token sequences that lie outside the empirical support of the training corpus. We evaluate the finite-sample performance of the proposed procedure using both synthetic and real-world language data. Results show that the proposed method consistently improves out-of-support prediction while maintaining competitive in-support performance, demonstrating that perturbation offers a practical route to language modeling.
展开完整摘要收起摘要↓
We introduce a simple yet powerful framework for training large language models. In contrast to the standard autoregressive next-token prediction based on an exact prefix, we propose a perturbation-based procedure that first transforms the prefix into a semantic neighbor and then conditions on this perturbed variant for next-token prediction. This yields a hierarchical model with a pre-post-additive noise structure. Within this framework, we develop a rigorous theory of extrapolability, namely, the capacity of a model class to make reliable predictions for token sequences that lie outside the empirical support of the training corpus. We evaluate the finite-sample performance of the proposed procedure using both synthetic and real-world language data. Results show that the proposed method consistently improves out-of-support prediction while maintaining competitive in-support performance, demonstrating that perturbation offers a practical route to language modeling.
Persistent external memory enables LLM agents to maintain context across sessions, yet its security properties remain formally uncharacterized. We formalize memory poisoning attacks on retrieval-augmented agents as a Stackelberg game with a unified evaluation framework spanning three attack classes with escalating access assumptions. Correcting an evaluation protocol inconsistency in the triggered-query specification of Chen et al. (2024), we show faithful evaluation increases measured attack success by $4\times$ (ASR-R: $0.25 \to 1.00$). Our primary contribution is MEMSAD (Semantic Anomaly Detection), a calibration-based defense grounded in a gradient coupling theorem: under encoder regularity, the anomaly score gradient and the retrieval objective gradient are provably identical, so any continuous perturbation that reduces detection risk necessarily degrades retrieval rank. This coupling yields a certified detection radius guaranteeing correct classification regardless of adversary strategy. We prove minimax optimality via Le Cam's method, showing any threshold detector requires $Ω(1/ρ^2)$ calibration samples and MEMSAD achieves this up to $\log(1/δ)$ factors. We further derive online regret bounds for rolling calibration at rate $O(σ^{2/3}Δ^{1/3})$, and formally characterize a discrete synonym-invariance loophole that marks the boundary of what continuous-space defenses can guarantee. Experiments on a $3 \times 5$ attack-defense matrix with bootstrap confidence intervals, Bonferroni-corrected hypothesis tests, and Clopper-Pearson validation ($n=1{,}000$) confirm: composite defenses achieve TPR $= 1.00$, FPR $= 0.00$ across all attacks, while synonym substitution evades detection at $Δ$ ASR-R $\approx 0$, exposing a gap existing embedding-based defenses cannot close.
展开完整摘要收起摘要↓
Persistent external memory enables LLM agents to maintain context across sessions, yet its security properties remain formally uncharacterized. We formalize memory poisoning attacks on retrieval-augmented agents as a Stackelberg game with a unified evaluation framework spanning three attack classes with escalating access assumptions. Correcting an evaluation protocol inconsistency in the triggered-query specification of Chen et al. (2024), we show faithful evaluation increases measured attack success by $4\times$ (ASR-R: $0.25 \to 1.00$). Our primary contribution is MEMSAD (Semantic Anomaly Detection), a calibration-based defense grounded in a gradient coupling theorem: under encoder regularity, the anomaly score gradient and the retrieval objective gradient are provably identical, so any continuous perturbation that reduces detection risk necessarily degrades retrieval rank. This coupling yields a certified detection radius guaranteeing correct classification regardless of adversary strategy. We prove minimax optimality via Le Cam's method, showing any threshold detector requires $Ω(1/ρ^2)$ calibration samples and MEMSAD achieves this up to $\log(1/δ)$ factors. We further derive online regret bounds for rolling calibration at rate $O(σ^{2/3}Δ^{1/3})$, and formally characterize a discrete synonym-invariance loophole that marks the boundary of what continuous-space defenses can guarantee. Experiments on a $3 \times 5$ attack-defense matrix with bootstrap confidence intervals, Bonferroni-corrected hypothesis tests, and Clopper-Pearson validation ($n=1{,}000$) confirm: composite defenses achieve TPR $= 1.00$, FPR $= 0.00$ across all attacks, while synonym substitution evades detection at $Δ$ ASR-R $\approx 0$, exposing a gap existing embedding-based defenses cannot close.
作者Severin Ye, Xiao Kong, Xiaopeng He, Guangsu Yan, Dongsuk Oh
Discharge summaries require extracting critical information from lengthy electronic health records (EHRs), a process that is labor-intensive when performed manually. Large language models (LLMs) can improve generation efficiency; however, they are prone to producing faithfulness hallucinations, statements that contradict source records, posing direct risks to patient safety. To address this, we present CuraView, a multi-agent framework for sentence-level detection and evidence-grounded explanation of faithfulness hallucinations in discharge summaries. CuraView constructs a GraphRAG-based knowledge graph from patient-level EHRs and implements a closed-loop generation-detection pipeline with sentence-level evidence retrieval and classification spanning four evidence grades from strong support to direct contradiction (E1-E4), yielding structured and interpretable evidence chains. We evaluate CuraView on a subset of 250 patients from the Discharge-Me benchmark, with 50 patients held out for testing. Our fine-tuned Qwen3-14B detection model achieves an F1 of 0.831 on the safety-critical E4 metric (90.9% recall, 76.5% precision) and an F1 of 0.823 on E3+E4, representing a 50.0% relative improvement over the base model and outperforming RAGTruth-style and QAGS-style baselines. These results demonstrate that evidence-chain-based graph retrieval verification substantially improves the factual reliability of clinical documentation, while simultaneously producing reusable annotated datasets for downstream model training and distillation.
展开完整摘要收起摘要↓
Discharge summaries require extracting critical information from lengthy electronic health records (EHRs), a process that is labor-intensive when performed manually. Large language models (LLMs) can improve generation efficiency; however, they are prone to producing faithfulness hallucinations, statements that contradict source records, posing direct risks to patient safety. To address this, we present CuraView, a multi-agent framework for sentence-level detection and evidence-grounded explanation of faithfulness hallucinations in discharge summaries. CuraView constructs a GraphRAG-based knowledge graph from patient-level EHRs and implements a closed-loop generation-detection pipeline with sentence-level evidence retrieval and classification spanning four evidence grades from strong support to direct contradiction (E1-E4), yielding structured and interpretable evidence chains. We evaluate CuraView on a subset of 250 patients from the Discharge-Me benchmark, with 50 patients held out for testing. Our fine-tuned Qwen3-14B detection model achieves an F1 of 0.831 on the safety-critical E4 metric (90.9% recall, 76.5% precision) and an F1 of 0.823 on E3+E4, representing a 50.0% relative improvement over the base model and outperforming RAGTruth-style and QAGS-style baselines. These results demonstrate that evidence-chain-based graph retrieval verification substantially improves the factual reliability of clinical documentation, while simultaneously producing reusable annotated datasets for downstream model training and distillation.
Workspace learning requires AI agents to identify, reason over, exploit, and update explicit and implicit dependencies among heterogeneous files in a worker's workspace, enabling them to complete both routine and advanced tasks effectively. Despite its importance, existing relevant benchmarks largely evaluate agents on pre-specified or synthesized files with limited real-world dependencies, leaving workspace-level evaluation underexplored. To this end, we introduce Workspace-Bench, a benchmark for evaluating AI agents on Workspace Learning invOlving Large-Scale File Dependencies. We construct realistic workspaces with 5 worker profiles, 74 file types, 20,476 files (up to 20GB) and curate 388 tasks, each with its own file dependency graph, evaluated across 7,399 total rubrics that require cross-file retrieval, contextual reasoning, and adaptive decision-making. We further provide Workspace-Bench-Lite, a 100-task subset that preserves the benchmark distribution while reducing evaluation costs by about 70%. We evaluate 4 popular agent harnesses and 7 foundation models. Experimental results show that current agents remain far from reliable workspace learning, where the best reaches only 68.7%, substantially below the human result of 80.7%, and the average performance across agents is only 47.4%.
展开完整摘要收起摘要↓
Workspace learning requires AI agents to identify, reason over, exploit, and update explicit and implicit dependencies among heterogeneous files in a worker's workspace, enabling them to complete both routine and advanced tasks effectively. Despite its importance, existing relevant benchmarks largely evaluate agents on pre-specified or synthesized files with limited real-world dependencies, leaving workspace-level evaluation underexplored. To this end, we introduce Workspace-Bench, a benchmark for evaluating AI agents on Workspace Learning invOlving Large-Scale File Dependencies. We construct realistic workspaces with 5 worker profiles, 74 file types, 20,476 files (up to 20GB) and curate 388 tasks, each with its own file dependency graph, evaluated across 7,399 total rubrics that require cross-file retrieval, contextual reasoning, and adaptive decision-making. We further provide Workspace-Bench-Lite, a 100-task subset that preserves the benchmark distribution while reducing evaluation costs by about 70%. We evaluate 4 popular agent harnesses and 7 foundation models. Experimental results show that current agents remain far from reliable workspace learning, where the best reaches only 68.7%, substantially below the human result of 80.7%, and the average performance across agents is only 47.4%.
作者Haoxi Li, Qinglin Hou, Jianfei Ma, Jinxiang Lai, Tao Han, Sikai Bai, Jingcai Guo, Jie Zhang, Song Guo
To navigate partially observable visual environments, recent VLM agents increasingly internalize world modeling capabilities into their policies via explicit CoT reasoning, enabling them to mentally simulate futures before acting. However, relying solely on passive reasoning over visited states is insufficient for sparse-reward tasks, as it lacks the epistemic drive to actively uncover the ``known unknown'' required for robust generalization. We ask: Can VLM agents actively find signals that challenge and refine their internal world model through curiosity-driven exploration? In this work, we propose GLANCE, a unified framework that bridges reasoning and exploration by grounding the agent's linguistic world model into the stable visual representations of an evolving target network. Crucially, GLANCE leverages the discrepancy between linguistic prediction and visual reality as an intrinsic curiosity signal within reinforcement learning, steering the agent to actively explore areas where its internal model is uncertain. Extensive experiments across a series of agentic tasks show the effectiveness of GLANCE, and demonstrate that aligning ``what the agent thinks'' with ``what the agent sees'' is key to solving complex or sparse agentic tasks.
展开完整摘要收起摘要↓
To navigate partially observable visual environments, recent VLM agents increasingly internalize world modeling capabilities into their policies via explicit CoT reasoning, enabling them to mentally simulate futures before acting. However, relying solely on passive reasoning over visited states is insufficient for sparse-reward tasks, as it lacks the epistemic drive to actively uncover the ``known unknown'' required for robust generalization. We ask: Can VLM agents actively find signals that challenge and refine their internal world model through curiosity-driven exploration? In this work, we propose GLANCE, a unified framework that bridges reasoning and exploration by grounding the agent's linguistic world model into the stable visual representations of an evolving target network. Crucially, GLANCE leverages the discrepancy between linguistic prediction and visual reality as an intrinsic curiosity signal within reinforcement learning, steering the agent to actively explore areas where its internal model is uncertain. Extensive experiments across a series of agentic tasks show the effectiveness of GLANCE, and demonstrate that aligning ``what the agent thinks'' with ``what the agent sees'' is key to solving complex or sparse agentic tasks.
作者Joseph Breda, Fadi Yousif, Beszel Hawkins, Marinela Cotoi, Miao Liu, Ray Luo, Po-Hsuan Cameron Chen, Mike Schaekermann, Samuel Schmidgall, Xin Liu, Girish Narayanswamy, Samuel Solomon, Maxwell A. Xu, Xiaoran Fan, Longfei Shangguan, Anran Wang, Bhavna Daryani, Buddy Herkenham, Cara Tan, Mark Malhotra, Shwetak Patel, John B. Hernandez, Quang Duong, Yun Liu, Zach Wasson, Dimitrios Antos, Bob Lou, Matthew Thompson, Jonathan Richina, Anupam Pathak, Nichole Young-Lin, Jake Sunshine, Daniel McDuff
Language models excel at diagnostic assessments on currated medical case-studies and vignettes, performing on par with, or better than, clinical professionals. However, existing studies focus on complex scenarios with rich context making it difficult to draw conclusions about how these systems perform for patients reporting symptoms in everyday life. We deployed SymptomAI, a set of conversational AI agents for end-to-end patient interviewing and differential diagnosis (DDx), via the Fitbit app in a study that randomized participants (N=13,917) to interact with five AI agents. This corpus captures diverse communication and a realistic distribution of illnesses from a real world population. A subset of 1,228 participants reported a clinician-provided diagnosis, and 517 of these were further evaluated by a panel of clinicians during over 250 hours of annotation. SymptomAI DDx were significantly more accurate (OR = 2.47, p < 0.001) than those from independent clinicians given the same dialogue in a blinded randomized comparison. Moreover, agentic strategies which conduct a dedicated symptom interview that elicit additional symptom information before providing a diagnosis, perform substantially better than baseline, user-guided conversations (p < 0.001). An auxiliary analysis on 1,509 conversations from a general US population panel validated that these results generalize beyond wearable device users. We used SymptomAI diagnoses as labels for all 13,917 participants to analyze over 500,000 days of wearable metrics across nearly 400 unique conditions. We identified strong associations between acute infections and physiological shifts (e.g., OR > 7 for influenza). While limited by self-reported ground truth, these results demonstrate the benefits of a dedicated and complete symptom interview compared to a user-guided symptom discussion, which is the default of most consumer LLMs.
展开完整摘要收起摘要↓
Language models excel at diagnostic assessments on currated medical case-studies and vignettes, performing on par with, or better than, clinical professionals. However, existing studies focus on complex scenarios with rich context making it difficult to draw conclusions about how these systems perform for patients reporting symptoms in everyday life. We deployed SymptomAI, a set of conversational AI agents for end-to-end patient interviewing and differential diagnosis (DDx), via the Fitbit app in a study that randomized participants (N=13,917) to interact with five AI agents. This corpus captures diverse communication and a realistic distribution of illnesses from a real world population. A subset of 1,228 participants reported a clinician-provided diagnosis, and 517 of these were further evaluated by a panel of clinicians during over 250 hours of annotation. SymptomAI DDx were significantly more accurate (OR = 2.47, p < 0.001) than those from independent clinicians given the same dialogue in a blinded randomized comparison. Moreover, agentic strategies which conduct a dedicated symptom interview that elicit additional symptom information before providing a diagnosis, perform substantially better than baseline, user-guided conversations (p < 0.001). An auxiliary analysis on 1,509 conversations from a general US population panel validated that these results generalize beyond wearable device users. We used SymptomAI diagnoses as labels for all 13,917 participants to analyze over 500,000 days of wearable metrics across nearly 400 unique conditions. We identified strong associations between acute infections and physiological shifts (e.g., OR > 7 for influenza). While limited by self-reported ground truth, these results demonstrate the benefits of a dedicated and complete symptom interview compared to a user-guided symptom discussion, which is the default of most consumer LLMs.
LLM-Agents have evolved into autonomous systems for complex task execution, with the SKILL.md specification emerging as a de facto standard for encapsulating agent capabilities. However, a critical bottleneck remains: different agent frameworks exhibit starkly different sensitivities to prompt formatting, causing up to 40% performance variation, yet nearly all skills exist as a single, format-agnostic Markdown version. Manual per-platform rewriting creates an unsustainable maintenance burden, while prior audits have found that over one third of community skills contain security vulnerabilities. To address this, we present SkCC, a compilation framework that introduces classical compiler design into agent skill development. At its core, SkIR - a strongly-typed intermediate representation - decouples skill semantics from platform-specific formatting, enabling portable deployment across heterogeneous agent frameworks. Around this IR, a compile-time Analyzer enforces security constraints via Anti-Skill Injection before deployment. Through a four-phase pipeline, SkCC reduces adaptation complexity from $O(m \times n)$ to $O(m + n)$. Experiments on SkillsBench demonstrate that compiled skills consistently outperform their original counterparts, improving pass rates from 21.1% to 33.3% on Claude Code and from 35.1% to 48.7% on Kimi CLI, while achieving sub-10ms compilation latency, a 94.8% proactive security trigger rate, and 10-46% runtime token savings across platforms.
展开完整摘要收起摘要↓
LLM-Agents have evolved into autonomous systems for complex task execution, with the SKILL.md specification emerging as a de facto standard for encapsulating agent capabilities. However, a critical bottleneck remains: different agent frameworks exhibit starkly different sensitivities to prompt formatting, causing up to 40% performance variation, yet nearly all skills exist as a single, format-agnostic Markdown version. Manual per-platform rewriting creates an unsustainable maintenance burden, while prior audits have found that over one third of community skills contain security vulnerabilities. To address this, we present SkCC, a compilation framework that introduces classical compiler design into agent skill development. At its core, SkIR - a strongly-typed intermediate representation - decouples skill semantics from platform-specific formatting, enabling portable deployment across heterogeneous agent frameworks. Around this IR, a compile-time Analyzer enforces security constraints via Anti-Skill Injection before deployment. Through a four-phase pipeline, SkCC reduces adaptation complexity from $O(m \times n)$ to $O(m + n)$. Experiments on SkillsBench demonstrate that compiled skills consistently outperform their original counterparts, improving pass rates from 21.1% to 33.3% on Claude Code and from 35.1% to 48.7% on Kimi CLI, while achieving sub-10ms compilation latency, a 94.8% proactive security trigger rate, and 10-46% runtime token savings across platforms.
Agent memory failures are silent: an LLM-based agent can produce a fluent response even when it fails to extract, retain, or retrieve the information needed across sessions. The write-manage-read loop describes the external pipeline of these systems but leaves open which internal computations implement each stage. Tracing internal feature circuits across the Qwen-3 family (0.6B--14B) and two memory frameworks (mem0 and A-MEM), we report three findings. First, control is detectable before content: routing circuitry is causally active at 0.6B, while content circuitry produces no detectable signal until 4B under our tracing setup, creating a deployment regime where small models route with apparent competence but silently fail at extraction and grounding. Second, within the content group, Write and Read share a late-layer hub that operates as a context-grounding substrate already present in the base model; only memory framing recruits a functional grounding direction on this substrate, and the hub transfers across both frameworks. Third, emergence does not imply steerability: although the content circuit becomes detectable at 4B, it becomes reliably steerable only at 8B, indicating that detection and intervention have distinct scale thresholds. As a practical implication, the feature-space separation between the two circuit groups enables per-operation failure localization at 76.2% accuracy without supervision, providing a stage-level diagnostic for otherwise silent agent-memory failures.
展开完整摘要收起摘要↓
Agent memory failures are silent: an LLM-based agent can produce a fluent response even when it fails to extract, retain, or retrieve the information needed across sessions. The write-manage-read loop describes the external pipeline of these systems but leaves open which internal computations implement each stage. Tracing internal feature circuits across the Qwen-3 family (0.6B--14B) and two memory frameworks (mem0 and A-MEM), we report three findings. First, control is detectable before content: routing circuitry is causally active at 0.6B, while content circuitry produces no detectable signal until 4B under our tracing setup, creating a deployment regime where small models route with apparent competence but silently fail at extraction and grounding. Second, within the content group, Write and Read share a late-layer hub that operates as a context-grounding substrate already present in the base model; only memory framing recruits a functional grounding direction on this substrate, and the hub transfers across both frameworks. Third, emergence does not imply steerability: although the content circuit becomes detectable at 4B, it becomes reliably steerable only at 8B, indicating that detection and intervention have distinct scale thresholds. As a practical implication, the feature-space separation between the two circuit groups enables per-operation failure localization at 76.2% accuracy without supervision, providing a stage-level diagnostic for otherwise silent agent-memory failures.
Long-running autonomous AI agents suffer from a well-documented memory coherence problem: tool-execution success rates degrade 14 percentage points over 72-hour operation windows due to four compounding failure modes in existing flat-file memory systems. We present MEMTIER, a tripartite memory architecture for the OpenClaw agent runtime that introduces a structured episodic JSONL store, a five-signal weighted retrieval engine, an attention-attributed cognitive weight update loop, an asynchronous consolidation daemon promoting episodic facts to a semantic tier, and a PPO-based policy framework for adapting retrieval weights (infrastructure validated; performance gains pending camera-ready). On the full 500-question LongMemEval-S benchmark (Wu et al., 2025), MEMTIER achieves Acc=0.382, F1=0.412 with Qwen2.5-7B on a consumer 6GB GPU - a +33 percentage point improvement over the full-context baseline (0.050 -> 0.382, i.e., 5% -> 38%). With DeepSeek-V4-Flash fact pre-population, single-session recall reaches 0.686-0.714, exceeding the paper's RAG BM25 GPT-4o baseline (0.560) on those categories. Temporal reasoning rises to 0.323 and multi-session synthesis to 0.173, demonstrating that structured semantic pre-population qualitatively changes what lightweight retrieval can achieve. All phases run locally on a consumer laptop with a 6GB GPU.
展开完整摘要收起摘要↓
Long-running autonomous AI agents suffer from a well-documented memory coherence problem: tool-execution success rates degrade 14 percentage points over 72-hour operation windows due to four compounding failure modes in existing flat-file memory systems. We present MEMTIER, a tripartite memory architecture for the OpenClaw agent runtime that introduces a structured episodic JSONL store, a five-signal weighted retrieval engine, an attention-attributed cognitive weight update loop, an asynchronous consolidation daemon promoting episodic facts to a semantic tier, and a PPO-based policy framework for adapting retrieval weights (infrastructure validated; performance gains pending camera-ready). On the full 500-question LongMemEval-S benchmark (Wu et al., 2025), MEMTIER achieves Acc=0.382, F1=0.412 with Qwen2.5-7B on a consumer 6GB GPU - a +33 percentage point improvement over the full-context baseline (0.050 -> 0.382, i.e., 5% -> 38%). With DeepSeek-V4-Flash fact pre-population, single-session recall reaches 0.686-0.714, exceeding the paper's RAG BM25 GPT-4o baseline (0.560) on those categories. Temporal reasoning rises to 0.323 and multi-session synthesis to 0.173, demonstrating that structured semantic pre-population qualitatively changes what lightweight retrieval can achieve. All phases run locally on a consumer laptop with a 6GB GPU.
作者Danny Hoang, Ryan Matthiessen, Christopher Miller, Nasir Mannan, Ruby ElKharboutly, David Gorsich, Matthew P. Castanier, Farhad Imani
High-precision CNC machining of free-form aerospace components requires bounded compensations informed by inspection, simulation, and process knowledge. Off-the-shelf large language model (LLM) assistants can generate text, but they do not reliably execute risk-constrained multi-step numerical workflows or provide auditable provenance for high-stakes decisions. We present multi-agent knowledge analysis (MAKA), a human-in-the-loop decision-support architecture that separates intent routing, tools-only quantitative analysis, knowledge graph retrieval, and critic-based verification that enforces physical plausibility, safety bounds, and provenance completeness before recommendations are surfaced for human approval. MAKA is instantiated on a Ti-6Al-4V rotor blade machining testbed by fusing virtual-machining path-tracking error fields, cutting-force and deflection simulations, and scan-based 3D inspection deviation maps from 16 blades. The analysis decomposes deviation into an evidence-linked pathing component, a drift-based wear proxy capturing systematic evolution across parts, a residual systematic compliance term, and a variability proxy for instability-aware escalation. In a three-level tool-orchestration benchmark (single-step through $\geq$3-step stateful sequences), MAKA improves successful tool execution by up to 87.5 percentage points relative to an unstructured single-model interaction pattern with identical tool access. Digital twin what-if studies show MAKA can coordinate traceable compensation candidates that reduce predicted surface deviation from order $10^{-2}$in to approximately $\pm 10^{-3}$in over most of the blade within the simulation environment, providing a pre-deployment verification signal for risk-aware human decision-making.
展开完整摘要收起摘要↓
High-precision CNC machining of free-form aerospace components requires bounded compensations informed by inspection, simulation, and process knowledge. Off-the-shelf large language model (LLM) assistants can generate text, but they do not reliably execute risk-constrained multi-step numerical workflows or provide auditable provenance for high-stakes decisions. We present multi-agent knowledge analysis (MAKA), a human-in-the-loop decision-support architecture that separates intent routing, tools-only quantitative analysis, knowledge graph retrieval, and critic-based verification that enforces physical plausibility, safety bounds, and provenance completeness before recommendations are surfaced for human approval. MAKA is instantiated on a Ti-6Al-4V rotor blade machining testbed by fusing virtual-machining path-tracking error fields, cutting-force and deflection simulations, and scan-based 3D inspection deviation maps from 16 blades. The analysis decomposes deviation into an evidence-linked pathing component, a drift-based wear proxy capturing systematic evolution across parts, a residual systematic compliance term, and a variability proxy for instability-aware escalation. In a three-level tool-orchestration benchmark (single-step through $\geq$3-step stateful sequences), MAKA improves successful tool execution by up to 87.5 percentage points relative to an unstructured single-model interaction pattern with identical tool access. Digital twin what-if studies show MAKA can coordinate traceable compensation candidates that reduce predicted surface deviation from order $10^{-2}$in to approximately $\pm 10^{-3}$in over most of the blade within the simulation environment, providing a pre-deployment verification signal for risk-aware human decision-making.
Tool-using agent systems powered by large language models (LLMs) are increasingly deployed across web, app, operating-system, and transactional environments. Yet existing safety benchmarks still emphasize explicit risks, potentially overstating a model's ability to judge deceptive or ambiguous trajectories. To address this gap, we introduce ROME (Red-team Orchestrated Multi-agent Evolution), a controlled benchmark-construction pipeline that rewrites known unsafe trajectories into more deceptive evaluation instances while preserving their underlying risk labels. Starting from 100 unsafe source trajectories, ROME produces 300 challenge instances spanning contextual ambiguity, implicit risks, and shortcut decision-making. Experiments show that these challenge sets substantially degrade safety-judgment performance, with hidden-risk cases remaining particularly non-trivial even for recent frontier models. We further study ARISE (Analogical Reasoning for Inference-time Safety Enhancement), a retrieval-guided inference-time enhancement that retrieves ReAct-style analogical safety trajectories from an external analogical base and injects them as structured reasoning exemplars. ARISE improves judgment quality without retraining, but is best viewed as a task-specific robustness enhancement rather than a standalone safety guarantee. Together, ROME and ARISE provide practical tools for stress-testing and improving agent safety judgment under deceptive distribution shifts.
展开完整摘要收起摘要↓
Tool-using agent systems powered by large language models (LLMs) are increasingly deployed across web, app, operating-system, and transactional environments. Yet existing safety benchmarks still emphasize explicit risks, potentially overstating a model's ability to judge deceptive or ambiguous trajectories. To address this gap, we introduce ROME (Red-team Orchestrated Multi-agent Evolution), a controlled benchmark-construction pipeline that rewrites known unsafe trajectories into more deceptive evaluation instances while preserving their underlying risk labels. Starting from 100 unsafe source trajectories, ROME produces 300 challenge instances spanning contextual ambiguity, implicit risks, and shortcut decision-making. Experiments show that these challenge sets substantially degrade safety-judgment performance, with hidden-risk cases remaining particularly non-trivial even for recent frontier models. We further study ARISE (Analogical Reasoning for Inference-time Safety Enhancement), a retrieval-guided inference-time enhancement that retrieves ReAct-style analogical safety trajectories from an external analogical base and injects them as structured reasoning exemplars. ARISE improves judgment quality without retraining, but is best viewed as a task-specific robustness enhancement rather than a standalone safety guarantee. Together, ROME and ARISE provide practical tools for stress-testing and improving agent safety judgment under deceptive distribution shifts.
Long-term personalized memory for LLM agents is challenging on resource-limited edge devices due to high storage costs and multimodal complexity. To address this, we propose ScrapMem, a framework that integrates multimodal data into "Scrapbook Page." ScrapMem introduces Optical Forgetting, an optical compression mechanism that progressively reduces the resolution of older memories, lowering storage cost while suppressing low-value details. To maintain semantic consistency, we construct an Episodic Memory Graph (EM-Graph) that organizes key events into a causal-temporal structure. Extensive experiments on the multimodal ATM-Bench showcase that ScrapMem provides three main benefits: (1) strong performance, achieving a new state-of-the-art with a 51.0% Joint@10 score; (2) high storage efficiency, reducing memory usage by up to 93% via optical forgetting; and (3) improved recall, increasing Recall@10 to 70.3% through structured aggregation. ScrapMem offers an effective and storage-efficient solution for on-device long-term memory in multimodal LLM agents.
展开完整摘要收起摘要↓
Long-term personalized memory for LLM agents is challenging on resource-limited edge devices due to high storage costs and multimodal complexity. To address this, we propose ScrapMem, a framework that integrates multimodal data into "Scrapbook Page." ScrapMem introduces Optical Forgetting, an optical compression mechanism that progressively reduces the resolution of older memories, lowering storage cost while suppressing low-value details. To maintain semantic consistency, we construct an Episodic Memory Graph (EM-Graph) that organizes key events into a causal-temporal structure. Extensive experiments on the multimodal ATM-Bench showcase that ScrapMem provides three main benefits: (1) strong performance, achieving a new state-of-the-art with a 51.0% Joint@10 score; (2) high storage efficiency, reducing memory usage by up to 93% via optical forgetting; and (3) improved recall, increasing Recall@10 to 70.3% through structured aggregation. ScrapMem offers an effective and storage-efficient solution for on-device long-term memory in multimodal LLM agents.
Retrieval-augmented generation systems often assume that one fixed retrieval pipeline is sufficient across heterogeneous tasks, yet factoid question answering, multi-hop reasoning, and scientific verification exhibit different retrieval preferences. We present Experience-RAG Skill, an agent-oriented pluggable retrieval orchestration layer positioned between the agent and the retriever pool. The proposed skill analyzes the current scene, consults an experience memory, selects an appropriate retrieval strategy, and returns structured evidence to the agent. Under a fixed candidate pool, Experience-RAG Skill achieves an overall nDCG@10 of 0.8924 on BeIR/nq, BeIR/hotpotqa, and BeIR/scifact, outperforming fixed single-retriever baselines and remaining competitive with Adaptive-RAG-style routing. The results suggest that retrieval strategy selection can be productively encapsulated as a reusable agent skill rather than being hard-coded in the upper workflow.
展开完整摘要收起摘要↓
Retrieval-augmented generation systems often assume that one fixed retrieval pipeline is sufficient across heterogeneous tasks, yet factoid question answering, multi-hop reasoning, and scientific verification exhibit different retrieval preferences. We present Experience-RAG Skill, an agent-oriented pluggable retrieval orchestration layer positioned between the agent and the retriever pool. The proposed skill analyzes the current scene, consults an experience memory, selects an appropriate retrieval strategy, and returns structured evidence to the agent. Under a fixed candidate pool, Experience-RAG Skill achieves an overall nDCG@10 of 0.8924 on BeIR/nq, BeIR/hotpotqa, and BeIR/scifact, outperforming fixed single-retriever baselines and remaining competitive with Adaptive-RAG-style routing. The results suggest that retrieval strategy selection can be productively encapsulated as a reusable agent skill rather than being hard-coded in the upper workflow.
作者Yuwen Du, Rui Ye, Shuo Tang, Keduan Huang, Xinyu Zhu, Yuzhu Cai, Siheng Chen
Deep search capabilities have become an indispensable competency for frontier Large Language Model (LLM) agents, yet their development remains dominated by industrial giants. The typical industry recipe involves a highly resource-intensive pipeline spanning pre-training, continual pre-training (CPT), supervised fine-tuning (SFT), and reinforcement learning (RL). In this report, we show that when fueled with informative and high-difficulty trajectories, a simple SFT approach could be surprisingly powerful for training frontier search agents. By introducing three simple data synthesis modifications: scaling knowledge graph size for richer exploration, expanding the tool set size for broader functionality, and strict low-step filtering, we establish a stronger baseline. Trained on merely 10.6k data points, our OpenSeeker-v2 achieves state-of-the-art performance across 4 benchmarks (30B-sized agents with ReAct paradigm): 46.0% on BrowseComp, 58.1% on BrowseComp-ZH, 34.6% on Humanity's Last Exam, and 78.0% on xbench, surpassing even Tongyi DeepResearch trained with heavy CPT+SFT+RL pipeline, which achieves 43.4%, 46.7%, 32.9%, and 75.0%, respectively. Notably, OpenSeeker-v2 represents the first state-of-the-art search agent within its model scale and paradigm to be developed by a purely academic team using only SFT. We are excited to open-source the OpenSeeker-v2 model weights and share our simple yet effective findings to make frontier search agent research more accessible to the community.
展开完整摘要收起摘要↓
Deep search capabilities have become an indispensable competency for frontier Large Language Model (LLM) agents, yet their development remains dominated by industrial giants. The typical industry recipe involves a highly resource-intensive pipeline spanning pre-training, continual pre-training (CPT), supervised fine-tuning (SFT), and reinforcement learning (RL). In this report, we show that when fueled with informative and high-difficulty trajectories, a simple SFT approach could be surprisingly powerful for training frontier search agents. By introducing three simple data synthesis modifications: scaling knowledge graph size for richer exploration, expanding the tool set size for broader functionality, and strict low-step filtering, we establish a stronger baseline. Trained on merely 10.6k data points, our OpenSeeker-v2 achieves state-of-the-art performance across 4 benchmarks (30B-sized agents with ReAct paradigm): 46.0% on BrowseComp, 58.1% on BrowseComp-ZH, 34.6% on Humanity's Last Exam, and 78.0% on xbench, surpassing even Tongyi DeepResearch trained with heavy CPT+SFT+RL pipeline, which achieves 43.4%, 46.7%, 32.9%, and 75.0%, respectively. Notably, OpenSeeker-v2 represents the first state-of-the-art search agent within its model scale and paradigm to be developed by a purely academic team using only SFT. We are excited to open-source the OpenSeeker-v2 model weights and share our simple yet effective findings to make frontier search agent research more accessible to the community.
作者Srinath Perera, Kaviru Hapuarachchi, Frank Leymann, Rania Khalaf
We present Robust Agent Compensation (RAC), a log-based recovery paradigm (providing a safety net) implemented through an architectural extension that can be applied to most Agent frameworks to support reliable executions (avoiding unintended side effects). Users can choose to enable RAC without changing their current agent code (e.g., LangGraph agents). The proposed approach can be implemented in most existing agent frameworks via their existing extension points. We present an implementation based on LangChain, demonstrate its viability through the $τ$-bench and REALM-Bench, and show that when solving complex problems, RAC is 1.5-8X or more better in both latency and token economy compared to state-of-the-art LLM-based recovery approaches.
展开完整摘要收起摘要↓
We present Robust Agent Compensation (RAC), a log-based recovery paradigm (providing a safety net) implemented through an architectural extension that can be applied to most Agent frameworks to support reliable executions (avoiding unintended side effects). Users can choose to enable RAC without changing their current agent code (e.g., LangGraph agents). The proposed approach can be implemented in most existing agent frameworks via their existing extension points. We present an implementation based on LangChain, demonstrate its viability through the $τ$-bench and REALM-Bench, and show that when solving complex problems, RAC is 1.5-8X or more better in both latency and token economy compared to state-of-the-art LLM-based recovery approaches.
Additive manufacturing (AM) continues to transform modern manufacturing by enabling flexible, on-demand production of complex geometries across diverse industries. Fused filament fabrication (FFF) has extended AM to laboratories, classrooms, and small production environments, but this accessibility shifts process-planning responsibility to users who may lack manufacturing expertise. A syntactically valid slicer profile can still encode thermally or geometrically harmful settings, and subtle G-code edits can alter extrusion, cooling, or adhesion before a print begins. Pre-print G-code screening catches accidental or adversarial machine-program errors before material or machine time is wasted. This paper proposes LLM-ADAM as a generalizable LLM framework for pre-print anomaly detection in AM. The framework decomposes the task into three roles: Extractor-LLM maps a G-code file to a structured process-parameter schema; Reference-LLM converts printer and material documentation into aligned operating ranges; and Judge-LLM interprets a deterministic deviation table and G-code evidence to decide whether a part is non-defective or belongs to an anomaly class. Printers, materials, and LLM backbones are interchangeable test conditions, not fixed assumptions. We evaluate the framework on an N=200 FFF G-code corpus spanning two desktop printer families, two materials, and five classes including non-defective, under-extrusion, over-extrusion, warping, and stringing. The best framework configuration reaches 87.5% accuracy, compared with 59.5% for the strongest engineered single-LLM baseline. The results show that structured decomposition, rather than backbone strength alone, is the dominant source of improvement, with defect classes identified at or near ceiling for leading configurations while residual errors concentrate on conservative false alarms for non-defective samples.
展开完整摘要收起摘要↓
Additive manufacturing (AM) continues to transform modern manufacturing by enabling flexible, on-demand production of complex geometries across diverse industries. Fused filament fabrication (FFF) has extended AM to laboratories, classrooms, and small production environments, but this accessibility shifts process-planning responsibility to users who may lack manufacturing expertise. A syntactically valid slicer profile can still encode thermally or geometrically harmful settings, and subtle G-code edits can alter extrusion, cooling, or adhesion before a print begins. Pre-print G-code screening catches accidental or adversarial machine-program errors before material or machine time is wasted. This paper proposes LLM-ADAM as a generalizable LLM framework for pre-print anomaly detection in AM. The framework decomposes the task into three roles: Extractor-LLM maps a G-code file to a structured process-parameter schema; Reference-LLM converts printer and material documentation into aligned operating ranges; and Judge-LLM interprets a deterministic deviation table and G-code evidence to decide whether a part is non-defective or belongs to an anomaly class. Printers, materials, and LLM backbones are interchangeable test conditions, not fixed assumptions. We evaluate the framework on an N=200 FFF G-code corpus spanning two desktop printer families, two materials, and five classes including non-defective, under-extrusion, over-extrusion, warping, and stringing. The best framework configuration reaches 87.5% accuracy, compared with 59.5% for the strongest engineered single-LLM baseline. The results show that structured decomposition, rather than backbone strength alone, is the dominant source of improvement, with defect classes identified at or near ceiling for leading configurations while residual errors concentrate on conservative false alarms for non-defective samples.
作者Shihao Weng, Yang Feng, Jinrui Zhang, Xiaofei Xie, Jiongchi Yu, Jia Liu
The rise of Large Language Model (LLM) agents, augmented with tool use, skills, and external knowledge, has introduced new security risks. Among them, prompt injection attacks, where adversaries embed malicious instructions into the agent workflow, have emerged as the primary threat. However, existing benchmarks and defenses are fundamentally limited as they assume context-insensitive settings in which the agent works under a fully specified user instruction, and the attacks are straightforward and context-independent. As a result, they fail to capture real-world deployments where agent behavior usually depends on dynamic context, not just the user prompt, and adversaries can adapt their attacks to different context. Similarly, existing defenses built on this narrow threat model overlook the nature of real-world agent delegation. In this paper, we present AgentLure, a benchmark that captures context-dependent tasks and context-aware prompt injection attacks. AgentLure spans four agentic domains and eight attack vectors across diverse attack surfaces. Our evaluation shows that existing defenses often struggle in this setting, yielding poor performance against such attacks in agentic systems. To address this limitation, we propose ARGUS, a defense mechanism that enforces provenance-aware decision auditing for LLM agents. ARGUS constructs an influence provenance graph to track how untrusted context propagates into agent decisions and verify whether a decision is justified by trustworthy evidence before execution. Our evaluation shows ARGUS reduces attack success rate to 3.8% while preserving 87.5% task utility, significantly outperforming existing defenses and remaining robust against adaptive white-box adversaries.
展开完整摘要收起摘要↓
The rise of Large Language Model (LLM) agents, augmented with tool use, skills, and external knowledge, has introduced new security risks. Among them, prompt injection attacks, where adversaries embed malicious instructions into the agent workflow, have emerged as the primary threat. However, existing benchmarks and defenses are fundamentally limited as they assume context-insensitive settings in which the agent works under a fully specified user instruction, and the attacks are straightforward and context-independent. As a result, they fail to capture real-world deployments where agent behavior usually depends on dynamic context, not just the user prompt, and adversaries can adapt their attacks to different context. Similarly, existing defenses built on this narrow threat model overlook the nature of real-world agent delegation. In this paper, we present AgentLure, a benchmark that captures context-dependent tasks and context-aware prompt injection attacks. AgentLure spans four agentic domains and eight attack vectors across diverse attack surfaces. Our evaluation shows that existing defenses often struggle in this setting, yielding poor performance against such attacks in agentic systems. To address this limitation, we propose ARGUS, a defense mechanism that enforces provenance-aware decision auditing for LLM agents. ARGUS constructs an influence provenance graph to track how untrusted context propagates into agent decisions and verify whether a decision is justified by trustworthy evidence before execution. Our evaluation shows ARGUS reduces attack success rate to 3.8% while preserving 87.5% task utility, significantly outperforming existing defenses and remaining robust against adaptive white-box adversaries.
作者Andrea Iannoli, Lorenzo Gigli, Luca Sciullo, Angelo Trotta, Marco Di Felice
Large Language Models (LLMs) are increasingly explored as high-level reasoning engines for cyber-physical systems, yet their application to real-time UAV swarm management remains challenging due to heterogeneous interfaces, limited grounding, and the need for long-running closed-loop execution. This paper presents a mission-agnostic, agent-enhanced LLM framework for UAV swarm control, where users express mission objectives in natural language and the system autonomously executes them through grounded, real-time interactions. The proposed architecture combines an LLM-based Agent Core with a Model Context Protocol (MCP) gateway and a Web-of-Drones abstraction based on W3C Web of Things (WoT) standards. By exposing drones, sensors, and services as standardized WoT Things, the framework enables structured tool-based interaction, continuous state observation, and safe actuation without relying on code generation. We evaluate the framework using ArduPilot-based simulation across four swarm missions and six state-of-the-art LLMs. Results show that, despite strong reasoning abilities, current general-purpose LLMs still struggle to achieve reliable execution - even for simple swarm tasks - when operating without explicit grounding and execution support. Task-specific planning tools and runtime guardrails substantially improve robustness, while token consumption alone is not indicative of execution quality or reliability.
展开完整摘要收起摘要↓
Large Language Models (LLMs) are increasingly explored as high-level reasoning engines for cyber-physical systems, yet their application to real-time UAV swarm management remains challenging due to heterogeneous interfaces, limited grounding, and the need for long-running closed-loop execution. This paper presents a mission-agnostic, agent-enhanced LLM framework for UAV swarm control, where users express mission objectives in natural language and the system autonomously executes them through grounded, real-time interactions. The proposed architecture combines an LLM-based Agent Core with a Model Context Protocol (MCP) gateway and a Web-of-Drones abstraction based on W3C Web of Things (WoT) standards. By exposing drones, sensors, and services as standardized WoT Things, the framework enables structured tool-based interaction, continuous state observation, and safe actuation without relying on code generation. We evaluate the framework using ArduPilot-based simulation across four swarm missions and six state-of-the-art LLMs. Results show that, despite strong reasoning abilities, current general-purpose LLMs still struggle to achieve reliable execution - even for simple swarm tasks - when operating without explicit grounding and execution support. Task-specific planning tools and runtime guardrails substantially improve robustness, while token consumption alone is not indicative of execution quality or reliability.
Group Relative Policy Optimization (GRPO) has recently shown strong performance in post-training large language models and vision-language models. It raises a question of whether the GRPO also significantly promotes the test-time adaptation (TTA) of vision language models. In this paper, we propose Group Relative Policy Optimization for Test-Time Adaptation (GRPO-TTA), which adapts GRPO to the TTA setting by reformulating class-specific prompt prediction as a group-wise policy optimization problem. Specifically, we construct output groups by sampling top-K class candidates from CLIP similarity distributions, enabling probability-driven optimization without access to ground-truth labels. Moreover, we design reward functions tailored to test-time adaptation, including alignment rewards and dispersion rewards, to guide effective visual encoder tuning. Extensive experiments across diverse benchmarks demonstrate that GRPO-TTA consistently outperforms existing test-time adaptation methods, with notably larger performance gains under natural distribution shifts.
展开完整摘要收起摘要↓
Group Relative Policy Optimization (GRPO) has recently shown strong performance in post-training large language models and vision-language models. It raises a question of whether the GRPO also significantly promotes the test-time adaptation (TTA) of vision language models. In this paper, we propose Group Relative Policy Optimization for Test-Time Adaptation (GRPO-TTA), which adapts GRPO to the TTA setting by reformulating class-specific prompt prediction as a group-wise policy optimization problem. Specifically, we construct output groups by sampling top-K class candidates from CLIP similarity distributions, enabling probability-driven optimization without access to ground-truth labels. Moreover, we design reward functions tailored to test-time adaptation, including alignment rewards and dispersion rewards, to guide effective visual encoder tuning. Extensive experiments across diverse benchmarks demonstrate that GRPO-TTA consistently outperforms existing test-time adaptation methods, with notably larger performance gains under natural distribution shifts.
The task of Long-tailed Class Incremental Learning (LT-CIL) addresses the sequential learning of new classes from datasets with imbalanced class distributions. This scenario intensifies the fundamental problem of catastrophic forgetting, inherent to continual learning, with the dual challenges of under-learning minority classes and overfitting majority classes. To tackle these combined issues, this paper proposes two main techniques. First, we introduce gradient consistency regularization, which leverages the moving average of gradients to suppress abrupt fluctuations and stabilize the training process. Second, we dynamically adjust the weight of the distillation loss by measuring the degree of class imbalance with normalized entropy. This adaptive weighting establishes an optimal balance between retaining old knowledge and acquiring new information. Experiments on the CIFAR-100-LT, ImageNetSubset-LT, and Food101-LT benchmarks show that our method achieves consistent accuracy improvements of up to 5.0%. Furthermore, we demonstrate dramatic gains in the challenging 'In-ordered' setting, where tasks progress from majority to minority classes, highlighting our method's robustness in mitigating forgetting under unfavorable learning dynamics. This enhanced performance is achieved without a significant increase in computational overhead, demonstrating the practicality of our framework.
展开完整摘要收起摘要↓
The task of Long-tailed Class Incremental Learning (LT-CIL) addresses the sequential learning of new classes from datasets with imbalanced class distributions. This scenario intensifies the fundamental problem of catastrophic forgetting, inherent to continual learning, with the dual challenges of under-learning minority classes and overfitting majority classes. To tackle these combined issues, this paper proposes two main techniques. First, we introduce gradient consistency regularization, which leverages the moving average of gradients to suppress abrupt fluctuations and stabilize the training process. Second, we dynamically adjust the weight of the distillation loss by measuring the degree of class imbalance with normalized entropy. This adaptive weighting establishes an optimal balance between retaining old knowledge and acquiring new information. Experiments on the CIFAR-100-LT, ImageNetSubset-LT, and Food101-LT benchmarks show that our method achieves consistent accuracy improvements of up to 5.0%. Furthermore, we demonstrate dramatic gains in the challenging 'In-ordered' setting, where tasks progress from majority to minority classes, highlighting our method's robustness in mitigating forgetting under unfavorable learning dynamics. This enhanced performance is achieved without a significant increase in computational overhead, demonstrating the practicality of our framework.
On-policy distillation (OPD) has recently emerged as an effective post-training paradigm for consolidating the capabilities of specialized expert models into a single student model. Despite its empirical success, the conditions under which OPD yields reliable improvement remain poorly understood. In this work, we identify two fundamental bottlenecks that limit effective OPD: insufficient exploration of informative states and unreliable teacher supervision for student rollouts. Building on this insight, we propose Uni-OPD, a unified OPD framework that generalizes across Large Language Models (LLMs) and Multimodal Large Language Models (MLLMs), centered on a dual-perspective optimization strategy. Specifically, from the student's perspective, we adopt two data balancing strategies to promote exploration of informative student-generated states during training. From the teacher's perspective, we show that reliable supervision hinges on whether aggregated token-level guidance remains order-consistent with the outcome reward. To this end, we develop an outcome-guided margin calibration mechanism to restore order consistency between correct and incorrect trajectories. We conduct extensive experiments on 5 domains and 16 benchmarks covering diverse settings, including single-teacher and multi-teacher distillation across LLMs and MLLMs, strong-to-weak distillation, and cross-modal distillation. Our results verify the effectiveness and versatility of Uni-OPD and provide practical insights into reliable OPD.
展开完整摘要收起摘要↓
On-policy distillation (OPD) has recently emerged as an effective post-training paradigm for consolidating the capabilities of specialized expert models into a single student model. Despite its empirical success, the conditions under which OPD yields reliable improvement remain poorly understood. In this work, we identify two fundamental bottlenecks that limit effective OPD: insufficient exploration of informative states and unreliable teacher supervision for student rollouts. Building on this insight, we propose Uni-OPD, a unified OPD framework that generalizes across Large Language Models (LLMs) and Multimodal Large Language Models (MLLMs), centered on a dual-perspective optimization strategy. Specifically, from the student's perspective, we adopt two data balancing strategies to promote exploration of informative student-generated states during training. From the teacher's perspective, we show that reliable supervision hinges on whether aggregated token-level guidance remains order-consistent with the outcome reward. To this end, we develop an outcome-guided margin calibration mechanism to restore order consistency between correct and incorrect trajectories. We conduct extensive experiments on 5 domains and 16 benchmarks covering diverse settings, including single-teacher and multi-teacher distillation across LLMs and MLLMs, strong-to-weak distillation, and cross-modal distillation. Our results verify the effectiveness and versatility of Uni-OPD and provide practical insights into reliable OPD.
作者Qichao Wang, Yunhong Lu, Hengyuan Cao, Junyi Zhang, Min Zhang
Dataset distillation enables efficient training by distilling the information of large-scale datasets into significantly smaller synthetic datasets. Diffusion based paradigms have emerged in recent years, offering novel perspectives for dataset distillation. However, they typically necessitate additional fine-tuning stages, and effective guidance mechanisms remain underexplored. To address these limitations, we rethink diffusion based dataset distillation and propose a Dual Matching Guided Diffusion (DMGD) framework, centered on efficient training-free guidance. We first establish Semantic Matching via conditional likelihood optimization, eliminating the need for auxiliary classifiers. Furthermore, we propose a dynamic guidance mechanism that enhances the diversity of synthetic data while maintaining semantic alignment. Simultaneously, we introduce an optimal transport (OT) based Distribution Matching approach to further align with the target distribution structure. To ensure efficiency, we develop two enhanced strategies for diffusion based framework: Distribution Approximate Matching and Greedy Progressive Matching. These strategies enable effective distribution matching guidance with minimal computational overhead. Experimental results on ImageNet-Woof, ImageNet-Nette, and ImageNet-1K demonstrate that our training-free approach achieves significant improvements, outperforming state-of-the-art (SOTA) methods requiring additional fine-tuning by average accuracy gains of 2.1%, 5.4%, and 2.4%, respectively.
展开完整摘要收起摘要↓
Dataset distillation enables efficient training by distilling the information of large-scale datasets into significantly smaller synthetic datasets. Diffusion based paradigms have emerged in recent years, offering novel perspectives for dataset distillation. However, they typically necessitate additional fine-tuning stages, and effective guidance mechanisms remain underexplored. To address these limitations, we rethink diffusion based dataset distillation and propose a Dual Matching Guided Diffusion (DMGD) framework, centered on efficient training-free guidance. We first establish Semantic Matching via conditional likelihood optimization, eliminating the need for auxiliary classifiers. Furthermore, we propose a dynamic guidance mechanism that enhances the diversity of synthetic data while maintaining semantic alignment. Simultaneously, we introduce an optimal transport (OT) based Distribution Matching approach to further align with the target distribution structure. To ensure efficiency, we develop two enhanced strategies for diffusion based framework: Distribution Approximate Matching and Greedy Progressive Matching. These strategies enable effective distribution matching guidance with minimal computational overhead. Experimental results on ImageNet-Woof, ImageNet-Nette, and ImageNet-1K demonstrate that our training-free approach achieves significant improvements, outperforming state-of-the-art (SOTA) methods requiring additional fine-tuning by average accuracy gains of 2.1%, 5.4%, and 2.4%, respectively.
作者Yue Shi, Peng Wang, Mingzhe Yu, Yunlong Zhao, Li Liu, Gareth D Hatton, Yan Lyu, Liangxiu Han
Diffusion-based voxel prior modelling is challenging for the reconstruction of large-scale 3D porous microstructures. Due to the demanding requirements for simultaneously modelling both the continuous pore morphology and the discrete pore-throat topology, the diffusion models require fully observed CT scans to provide topology-faithful priors, which results in an inherent trade-off among throughput, topological fidelity, and field of view in practical industrial applications. We propose GeoTopoDiff, a graph diffusion-based framework for reconstructing 3D porous microstructures from sparse CT slices. GeoTopoDiff transfers the learning of diffusion priors from a voxel-based space to a mixed graph state space, which simultaneously encompasses continuous pore geometry and discrete pore-throat topology. A topology-aware partial graph prior from sparsely observed CT slices is introduced to constrain the reverse denoising process. Experiments on anisotropic PTFE and Fontainebleau sandstone show that GeoTopoDiff reduces morphology-related errors by 19.8% and topology-sensitive transport errors by 36.5% on average. Our findings suggest that the mixed graph state space promotes the diffusion denoising process to reduce posterior uncertainty under a sparse observations. All models and code have been made publicly available to facilitate the exploration of diffusion models in the field of 3D porous microstructures simulation.
展开完整摘要收起摘要↓
Diffusion-based voxel prior modelling is challenging for the reconstruction of large-scale 3D porous microstructures. Due to the demanding requirements for simultaneously modelling both the continuous pore morphology and the discrete pore-throat topology, the diffusion models require fully observed CT scans to provide topology-faithful priors, which results in an inherent trade-off among throughput, topological fidelity, and field of view in practical industrial applications. We propose GeoTopoDiff, a graph diffusion-based framework for reconstructing 3D porous microstructures from sparse CT slices. GeoTopoDiff transfers the learning of diffusion priors from a voxel-based space to a mixed graph state space, which simultaneously encompasses continuous pore geometry and discrete pore-throat topology. A topology-aware partial graph prior from sparsely observed CT slices is introduced to constrain the reverse denoising process. Experiments on anisotropic PTFE and Fontainebleau sandstone show that GeoTopoDiff reduces morphology-related errors by 19.8% and topology-sensitive transport errors by 36.5% on average. Our findings suggest that the mixed graph state space promotes the diffusion denoising process to reduce posterior uncertainty under a sparse observations. All models and code have been made publicly available to facilitate the exploration of diffusion models in the field of 3D porous microstructures simulation.
We propose a unified, few-step generative modeling framework based on cumulative flow maps for long-range transport in probability space, inspired by flow-map techniques for physical transport and dynamics. At its core is a cumulative-flow abstraction that connects local, instantaneous updates with finite-time transport, enabling generative models to reason about global state transitions. This perspective yields a unified few-step framework built on cumulative transport and \revise{cumulative} parameterization that applies broadly to existing diffusion- and flow-based models without being tied to a specific prediction \revise{instantiation}. Our formulation supports few-step and even one-step generation while preserving synthesis quality, requiring only minimal changes to time embeddings and training objectives, and no increase in model capacity. We demonstrate its effectiveness across diverse tasks, including image generation, geometric distribution modeling, joint prediction, and SDF generation, with reduced inference cost.
展开完整摘要收起摘要↓
We propose a unified, few-step generative modeling framework based on cumulative flow maps for long-range transport in probability space, inspired by flow-map techniques for physical transport and dynamics. At its core is a cumulative-flow abstraction that connects local, instantaneous updates with finite-time transport, enabling generative models to reason about global state transitions. This perspective yields a unified few-step framework built on cumulative transport and \revise{cumulative} parameterization that applies broadly to existing diffusion- and flow-based models without being tied to a specific prediction \revise{instantiation}. Our formulation supports few-step and even one-step generation while preserving synthesis quality, requiring only minimal changes to time embeddings and training objectives, and no increase in model capacity. We demonstrate its effectiveness across diverse tasks, including image generation, geometric distribution modeling, joint prediction, and SDF generation, with reduced inference cost.
Multi-agent LLM systems on edge devices need to hand off latent context efficiently, but the practical choices today are expensive re-prefill or full-precision KV transfer. We study QKVShare, a framework for quantized KV-cache handoff between agents that combines token-level mixed-precision allocation, a self-contained CacheCard representation, and a HuggingFace-compatible cache injection path. Our current results support a narrower but clearer story than the original draft: on 150 GSM8K problems with Llama-3.1-8B-Instruct, adaptive quantization remains competitive under repeated handoff and shows its clearest gains against uniform quantization in deeper-hop, higher budget settings; for handoff latency, the QKVShare path reduces TTFT relative to full re prefill at every tested context, from 130.7 ms vs. 150.2 ms at nominal 1K context to 397.1 ms vs. 1029.7 ms at nominal 8K context;. Stage timing shows that post-injection generation, not card creation, dominates the current QKVShare latency path. These results position quantized KV handoff as a promising on-device systems direction while also highlighting the need for stronger controller ablations and apples-to-apples runtime comparisons.
展开完整摘要收起摘要↓
Multi-agent LLM systems on edge devices need to hand off latent context efficiently, but the practical choices today are expensive re-prefill or full-precision KV transfer. We study QKVShare, a framework for quantized KV-cache handoff between agents that combines token-level mixed-precision allocation, a self-contained CacheCard representation, and a HuggingFace-compatible cache injection path. Our current results support a narrower but clearer story than the original draft: on 150 GSM8K problems with Llama-3.1-8B-Instruct, adaptive quantization remains competitive under repeated handoff and shows its clearest gains against uniform quantization in deeper-hop, higher budget settings; for handoff latency, the QKVShare path reduces TTFT relative to full re prefill at every tested context, from 130.7 ms vs. 150.2 ms at nominal 1K context to 397.1 ms vs. 1029.7 ms at nominal 8K context;. Stage timing shows that post-injection generation, not card creation, dominates the current QKVShare latency path. These results position quantized KV handoff as a promising on-device systems direction while also highlighting the need for stronger controller ablations and apples-to-apples runtime comparisons.