每天知道
你的领域
出了什么

Streaming reconstruction of Free-Viewpoint Videos (FVVs) supports immersive Internet of Things (IoT) services, such as telepresence and digital twin visualization. Existing methods suffer from high per-frame optimization time and large storage footprints, limiting deployment on resource-constrained Edge-IoT devices. To address these challenges, we propose Structured Sparse Gaussian Streaming (S$^2$GS), an FVV reconstruction framework that exploits structure-aware temporal sparsity to selectively update Gaussian residuals, enabling efficient streaming without compromising visual fidelity. In the spatial domain, a streaming octree hierarchically organizes Gaussian residuals, capturing spatial correlations that guide residual updates. In the temporal domain, a structured gating mechanism, comprising hierarchical feature propagation (HFP) and Gumbel-Sigmoid sampling, converts hierarchical dynamic cues into sparse residual update decisions under differentiable optimization. A multi-level discrete scheme is further adopted to provide fine-grained control over residual updates while preserving intricate dynamic details. Extensive experiments across consumer GPUs, industrial edge IoT devices, and a physical telepresence testbed demonstrate that S$^2$GS consistently reduces per-frame optimization time and storage footprint while maintaining competitive visual quality. Compared with QUEEN, S$^2$GS reduces per-frame optimization time by 59% and storage costs by 85% on an RTX 4090 GPU. On the Jetson AGX Orin, S$^2$GS delivers the highest rendering throughput (60+ FPS) and the lowest energy consumption among the evaluated methods, demonstrating its potential for deployment in resource-constrained systems.

Choosing tensor-parallel (TP) degrees and replica counts for an LLM serving fleet is difficult because performance is not monotonic in TP and the feasible choice can change with load. Exhaustive profiling resolves this uncertainty, but measures many configurations that do not affect the final resource allocation. We present FleetSieve, which selects measurements according to their expected effect on a resource-coupled, SLO-aware fleet decision. FleetSieve models capacity and tail latency jointly, compares conservative and optimistic allocations, and stops when their remaining decision gap is below a specified tolerance. On a fixed H100 measurement grid for a 31B-parameter open-weight model, FleetSieve reaches the oracle aggregate decision using 22,200 GPU-seconds, 6.9% less than uniform random profiling in the fixed comparison. Across 200 random reveal orders, its mean saving over random profiling is 5.4% (95% bootstrap CI: 3.5-7.2%). The fixed-comparison saving is 21.5% for Chat, while FleetSieve does not use the fewest GPU-seconds for Code. Joint capacity and tail modeling also avoids selecting a configuration whose 46.4-second completion p99 violates a 30-second SLO. In a 16-GPU allocation, an incorrect sparse-profile decision loses up to 1.93 requests/s and 12.4 percentage points of max-min fulfillment. Boundary repeats and BurstGPT measurements support the observed load-dependent tail-latency mechanism.

Attention directly derives normalized information flow from pairwise scores. We introduce Relation, an alternative token-mixing primitive that first organizes pairwise evidence into explicit Self and Exchange relations and derives information flow afterward. This relational organization gives rise to Full Relation, FlashRelation, Linear Relation, Hybrid Relation, and a KV-style Relation Cache. Across matched decoder-only models at approximately 10M, 30M, and 100M parameters, Full Relation achieves lower final validation NLL than MHA at all three scales. In a fixed-context reference benchmark, FlashRelation is 3.60-4.41x faster than the materialized Full Relation implementation. Across scale-matched production workloads, it reaches 76.4-84.9% of PyTorch FlashAttention throughput while executing the Full Relation operator. Hybrid Relation uses 75% Linear Relation layers and achieves strong language-modeling quality. These results support a relation-first view of token mixing: ask Self, ask Others, then let Flow follow Relation.

Following a target speaker in a noisy environment, commonly known as the cocktail party problem, remains particularly challenging for cochlear implant (CI) users. Recent studies have explored EEG-based auditory attention decoding (AAD) using neural networks to enhance hearing assistance. This paper presents a resource-efficient ASIC for real-time EEG-based auditory attention decoding by integrating a quantized CNN inference engine and a Pearson-correlation classifier. The proposed architecture employs streaming execution, on-chip buffering, and memory-efficient dataflow to reduce hardware cost while maintaining real-time performance. The proposed ASIC has been fully implemented in GF22FDX 22-nm CMOS technology, occupying a total silicon area of 2.09 mm$^2$(1264$μ$m x 1654$μ$m), with the CNN inference engine and streaming classification engine requiring only 0.076 mm$^2$. Operating at a core voltage of 0.55 V, the design achieves a power consumption of 0.4941 mW and an inference latency of 7.34 ms, providing an energy-efficient hardware platform for EEG-based auditory attention decoding in hearing-assistance applications.

Modern optimizing compilers rely on heuristic search algorithms for NP-hard optimization problems, which can result in poor generated-code performance and long or unpredictable compile times. These are considered bugs by users, but verified compilers rarely reason beyond semantic preservation. We propose verifying performance and compile time properties of compiler passes. As a proof-of-concept, we formulate inline expansion using a cost model estimating instruction-cache performance. We mechanize this in Rocq, prove semantic preservation of the inlining transformation, and verify the algorithm's monotone improvement, convergence-time bound, and performance bounds for intermediate and final solutions.

Small language models are usually built like large ones and then squeezed onto a CPU afterwards. We did the opposite: we fixed the target first, one user, one token at a time, 4-bit weights, ordinary CPU, and chose the architecture to suit it. The result keeps full attention in only 6 of its 18 blocks. The other 12 use short convolutions whose memory is two timesteps wide no matter how long the conversation gets, so two thirds of the network never re-reads a growing cache. Trained from scratch on 59.9B tokens, the model scores 47.31 on a five-task benchmark against a bar of 42.20 that was fixed before training began. It beats GPT-2 124M, Pythia-160M, OPT-125M and GPT-neo-125M, all trained on three to six times more data, and exceeds MobileLLM-125M's published score despite that model seeing a trillion tokens. Validation bits-per-byte is 0.8685. To check the architecture rather than the training recipe, we trained a conventional all-attention model of the same size on the same data, and wrote down the winning condition before scoring either. The hybrid won the chosen quality metric by 0.81%, matched it on downstream tasks, produced a 6.3% smaller 4-bit file, and decoded 1.76x faster at 2048 tokens of context, 2.08x against an external model of similar size. In every measurement the speed advantage is near zero at an empty context and grows with length, which is what the mechanism predicts and what a merely leaner model would not show. A simple bandwidth calculation predicts only 1.17x, so memory volume alone does not explain the gap. We also report what did not work: an unmitigated 4-bit quality cost, roughly half the convolution channels ending up inert and impossible to remove, and a vocabulary larger than this model size warrants.

The entire ecosystem of open-source language models effectively relies on a single platform. What if this platform was forced to shut down tomorrow? Implementing and maintaining efficient model definitions and translating them between different training and inference regimes is a resource-heavy task that severely limits model efficiency and portability, hindering both scaling and deployment. Here, we present Axon, a strongly typed domain-specific language with Haskell-like syntax, that enables a write-once, run everywhere paradigm for LLM architectures. By basing collaboration on a language specification rather than a specific framework's vision, Axon fosters open cooperation and empowers researchers to implement highly specialized architectures without giving up optimization infrastructure or accepting deployment lock-in. Axon allows for concise, auditable specifications that can be automatically compiled to standalone implementations for leading frameworks: PyTorch, PyTorch with Triton, JAX, MLX and vLLM. In 467 inference benchmarking experiments on models ranging from 135M to 32B parameters, we demonstrate median speedups of 7% on PyTorch, 12% on PyTorch with Triton, 91% on JAX, and 107% on MLX, compared to the reference implementations from Transformers. When deployed as native vLLM architectures with PagedAttention and KV-cache, Axon models achieve a 58% median speedup over Transformers implementations.

Modern GPUs increasingly integrate Tensor Cores into the execution pipeline. Although aggregate tensor throughput continues to grow, aided by an operand supply that has evolved from register-based in Ampere to redundancy-free, memory-based in Hopper and Blackwell, efficiently orchestrating the complete tensor compute pipeline for the modern AI workloads remains challenging. We identify the fundamental bottlenecks as fixed parallelism and coarse-grained scheduling, both of which are exposed by modern AI workloads that interleave diverse non-GEMM operations with GEMM. To orchestrate tensor computation efficiently, we propose FIBER, a new architecture that extends the GPU SIMT (single instruction, multiple thread) model. Its basic execution instance, the fiber, is decoupled from private register ownership, carrying only minimal control state while accessing an SM's registers through a shared view. This enables dynamic parallelism scaling, fine-grained register-level dataflow scheduling, and offers a redundancy-free alternative for matrix operand supply. We extend the ISA, microarchitecture, and compiler to realize shared-register addressing, conflict-free operand delivery, and fiber-based program mapping. Under a typical mixed-precision LLM serving scenario, FIBER achieves a 2.25x end-to-end speedup on Ampere (1.15x for the original FP16 computation), with 1.8x and 2.09x on Hopper and Blackwell respectively, and kernel-level gains up to 2.49x.

Retrieval-augmented generation (RAG) improves language-model responses by grounding generation in external passages, which comes with overhead: retrieved context lengthens the prompt, increasing prefill work, KV-cache footprint, memory traffic, latency, and energy. Context compression offers a natural remedy by pruning retrieved text before generation. However, state-of-the-art context-compression methods are typically used with a fixed compression budget, or with the rate selected offline and then applied at inference time. This static view ignores both workload variation and the live state of the edge device. On an edge SoC, compression is not free: the compressor itself runs on the same SoC and consumes latency and energy that can offset any generation savings. This paper proposes a vision for telemetry-informed adaptive compression in edge RAG, grounded in experimental evidence. We characterize the compression tradeoff on the NVIDIA Jetson AGX Thor using Llama and Qwen generators, Natural Questions and HotpotQA datasets, and LLMLingua-2 compression. Our measurements show that generation dominates the RAG budget for larger models, reaching roughly 90% of per-query latency and 91% of GPU energy for 7B-8B generators. Exploring the impact of the compression rate reveals an adaptive operating region: mild compression can miss energy opportunities, and overly aggressive compression can hurt inference quality. Intermediate compression can reduce GPU energy by up to 53.2%, and SoC energy by up to 48.2%, with negligible quality loss. We argue for runtime policies that dynamically manage compression, guided by workload features and edge telemetry.

Agentic language models repeatedly encode tool and skill schemas that recur across requests in different combinations and orders, preventing standard prefix caching from reusing their key--value (KV) states. We introduce ReCache, a framework for independently caching resource representations while reducing their inference-time computational and memory overhead. Resource-wise attention removes cross-resource interactions and assigns resource-local positions, producing composition-invariant KV blocks. ReCache then restricts resource visibility to contribution-selected layer--KV-head-group routes and retains only invocation-critical fields through structural and semantic pruning. We evaluate ReCache on a benchmark assembled from seven public tool- and skill-use datasets, including resource-disjoint tests. Resource-wise attention matches dense invocation performance (82.3% versus 82.4% Inv-F1) while providing a 3.655$\times$ time-to-first-token speedup. The complete framework reduces allocated KV-tensor memory by 92.43% and accelerates attention by 1.423$\times$. These results show that separating reusable schema encoding from selective resource access substantially reduces agentic inference costs with limited effectiveness loss. The code is available at https://github.com/EIT-NLP/ReCache.

Prefix caching avoids prefill only when a repeated request returns to a server that still holds the prefix KV. Cache-blind balancing disperses that reuse; fixed affinity preserves it but can overload a server. CacheRoute resolves this tradeoff with a periodic routing plan. It admits high-rate keys to a stable warm set and places their assignments by expected load. Hot keys may use more than one destination, although every key in our primary semi-synthetic aggregate uses exactly one. On Llama-3.3-70B in fp8 across 60 H100 GPUs, CacheRoute sustains 176+/-11 QPS at a 3.5-s p99 SLO, 2.3x the strongest of five baselines. Served KV-cache hit rate rises from 64.1+/-1.3% under cache-blind balancing to 93.2+/-0.5%. A second semi-synthetic aggregate and controlled 8B and burst experiments separate the effects of affinity and placement. Two 32B workloads provide the counterexamples: when affinity recovers too little KV work, its residual load skew reduces or erases the improvement. We therefore recommend gating any deployment with a shadow replay rather than enabling affinity from workload statistics alone.

Long-context modeling is a pivotal capability for Large Language Models, yet the quadratic complexity of attention remains a critical bottleneck, particularly during the compute-intensive prefilling phase. Our previous work, FlashPrefill, mitigates this cost through instantaneous pattern discovery and max-based dynamic thresholding; however, it remains an algorithmic prototype that is still distant from production deployment. In this paper, we present FlashPrefill V2, which evolves FlashPrefill from a prototype toward practical long-context serving along three dimensions. First, we introduce a mean correction term that effectively suppresses the approximation error, keeping performance degradation manageable even at extreme sparsity levels. Second, we redesign the sparse attention operator with PackGQA memory access, warp specialization, and pingpong pipelining, fully aligning with the latest FlashAttention-3/4 implementations and supporting FP8 inference to meet practical quantization requirements. Third, FlashPrefill V2 natively supports paged KV cache and continuous batching, allowing integration as an attention backend in modern inference frameworks such as SGLang. Extensive evaluations on NVIDIA H20 GPUs---among the most widely deployed inference accelerators---demonstrate that FlashPrefill V2 delivers up to 47.26x and 27.19x speedups over FlashAttention-2 at 128K context length under FP8 and BF16 precision, respectively, and, in FP8, still achieves a 30.49x speedup against an FA3/4-aligned dense baseline.

The typical training process of a multimodal large language model (MLLM) involves adapting both the language model backbone and the projector between the backbone and a modality-specific encoder. We ask whether fine-tuning the backbone of an MLLM is necessary to adapt it to a new modality. Through experiments on 3D MLLMs, we find that training only the projector is sufficient to achieve strong multimodal performance relative to existing baseline models and our jointly trained MLLMs with the same encoder and backbone. We also show that joint training leads to undesirable drift in existing capabilities of the language model, which projector-only training avoids by definition. Furthermore, projector-only training has approximately twice the training sample throughput of joint training. We validate our findings across different language model backbones via 3D classification and captioning benchmarks as well as standard benchmarks evaluating language, vision, and spatial reasoning capabilities.

Multimodal sentiment analysis aims to infer affective states by integrating language, visual, and acoustic cues. However, real-world multimodal inputs are often incomplete or corrupted, which can weaken cross-modal complementarity and introduce misleading information into downstream fusion. Existing proxy-based methods for incomplete MSA commonly rely on one-shot proxy construction to compensate for degraded language information, but the generated proxy may be coarse or unreliable at initialization. Prematurely injecting such a proxy into multimodal reasoning can propagate initial errors and compromise sentiment prediction. To address this limitation, we propose an iterative proxy correction framework for robust incomplete MSA. Our method constructs a language-oriented proxy from non-language modalities and progressively refines it under multimodal context through gated residual correction. The corrected proxy is then adaptively fused with the observed language representation according to an estimated language reliability score, allowing the model to balance proxy-based compensation and trustworthy linguistic evidence. In addition, we introduce a stage-wise latent correction objective that uses the complete language representation as a training-time semantic anchor to stabilize the proxy refinement trajectory. Extensive experiments on MOSI, MOSEI, and SIMS under diverse missing-modality settings demonstrate that the proposed framework consistently outperforms competitive baselines and achieves robust sentiment prediction under incomplete inputs.

Causal Bayesian networks (CBNs) and structural causal models (SCMs) are the dominant frameworks for graphical causal reasoning, but they cannot adequately represent all real-world causal systems. In particular, systems at equilibrium---where feedback mechanisms create cyclic causal dependencies---can exhibit causal semantics that are fundamentally incompatible with these frameworks: different interventions that enforce the same variable value may have different effects, rendering the standard ``perfect intervention'' do($X = x$) ambiguous. We propose bipartite graphical causal models (BGCMs), in which the structure of a system of equations is encoded by a bipartite graph with variable and equation nodes. In this framework, a hard intervention do($f_j : X_v = ξ_v$) specifies which equation is replaced, which variable is targeted, and at what value---resolving the ambiguity of the standard notion. We demonstrate, through a detailed case study of a physical system, that this representation naturally corresponds to distinct real-world interventions. We formulate a Markov property in terms of a new graphical separation criterion (B-separation) that exploits the functional determinism inherent in the equations, and we extend it to settings with non-random inputs. We show how this gives rise to a do-calculus for reasoning about domain invariances. BGCMs strictly generalize CBNs and SCMs while retaining the ability to perform graphical causal reasoning.

Where does the novelty a base language model produces with no task come from, and what can an LLM judge of a long stream actually see? We dismantle a cognitively inspired generation loop over 24 conditions on three base models. Most of its effect lives in one operation: a new subject injected every few hundred tokens (an interruption) into a stream whose literal repetition is damped (habituation). We judge windows of generated text only, with the premise as the unit (n=10) and a judge measured for repeatability, against a second judge family and against human readers. Under that protocol the interruption raises judged surprise by 1.2 to 1.4 points and connection by 0.8 over habituation alone. A connective that asks for continuity hurts; a bare paragraph break adds nothing detectable on fresh text; a reset context does at least as well as a kept one; and a pre-registered replication on new premises confirms the primary contrast. Three things the window judge could not see changed the first version of this study, and we think they are of general use. The judge scores the experimenter's injected sentence as the model's own. A fixed rotation of injected sentences makes the model replay its earlier segments from beyond the judge's horizon, and the judge scores the replay as surprise and connection (65-80% of post-interruption windows at periods 150-300). And the local gains do not compose: no arm produces an integrated document. The salience monitor, the in-loop judge, memory across interruptions and a judge-gated Review run with a gate that opens add nothing. On a problem with a verifier (online bin packing), the interruption multiplies valid, distinct candidate heuristics three- to fourfold without raising the quality of the best. We report an evaluation protocol for long generation and a controlled characterization of a simple intervention, not a mechanism of creativity.

Discovering materials with desirable properties often requires searching large candidate spaces while experimental or computational evaluations remain costly. Active learning addresses this challenge by using previous observations to select which candidate to evaluate next, typically through probabilistic surrogate models. We investigate whether open-weight large language models (LLMs) can serve as standalone acquisition policies in this setting. We evaluate five LLMs across four retrospective finite-pool materials optimization tasks under different candidate-presentation strategies and compare them with random selection and conventional Gaussian-process methods. LLM policies generally reach the global optimum in fewer iterations than random selection, indicating that they provide a useful acquisition signal without task-specific training. Their performance relative to Gaussian-process methods is mixed: conventional acquisition performs better on most tasks, while LLMs match or outperform it in some settings. Performance varies substantially across tasks, models, initializations, and candidate presentations, with no LLM approach performing best across all tasks. Overall, open-weight LLMs show potential as acquisition policies for finite-pool materials search, although their reliability remains sensitive to the task and to how candidates and scientific context are presented.

Hidden chain-of-thought (CoT) traces, especially those from frontier proprietary large reasoning models (LRMs), are valuable model assets. Yet whether these hidden CoTs can be directly extracted from black-box models remains largely unexplored. In this work, we systematically study whether hidden CoTs can be extracted near-verbatim from black-box LRMs through API interactions. We identify a previously overlooked reasoning replay surface between tool calls and develop EchoCoT, a multi-step attack that iteratively extracts hidden CoTs using API-returned fidelity signals. We further develop an LLM-based optimization framework that automatically searches for an effective universal injection trajectory across various datasets. We evaluate EchoCoT on three open-source and five frontier proprietary LRMs. On open-source LRMs, EchoCoT achieves up to 66.4% near-verbatim extraction success, with the extracted trace length within 10% of the target and at least 90% of tokens exactly matching the target CoT. The same injection trajectory also generalizes to unseen datasets, achieving up to 80% extraction success under the same criterion. For tested frontier proprietary LRMs, a substantial fraction of extracted CoTs closely align with provider-reported reasoning lengths and available CoT summaries. EchoCoT can also extract very long CoTs: on Gemini-2.5, it extracts 33,463 tokens from a 32,948-token target. These results establish hidden-CoT extraction as a practical security risk and highlight the need to better protect hidden CoT assets.

Large Language Models (LLMs) are increasingly deployed in high-stakes applications, yet their tendency to generate toxic, harmful, or policy-violating content poses significant risks. Detecting these unsafe outputs efficiently in a black-box manner remains an open challenge. In this paper, we extend a recently proposed dynamical systems framework designed for hallucination detection to LLM safety classification. By projecting both prompts and responses into high-dimensional embedding spaces and fitting separate Koopman-based predictive models for safe and unsafe regimes, we classify new outputs using a new differential residual score that compares prediction errors of the safe and unsafe regimes. A key contribution is the incorporation of the prompt and response embedding dynamics, yielding fitted Koopman operators that capture crucial interaction patterns. We evaluate our black-box method across three safety benchmarks using three embedding models. Our results show that incorporating prompt embeddings yields consistent improvements, particularly for interaction-dependent violations when paired with causal decoders (e.g., in Llama-3), while response-only violations benefit more from dense semantic embedding representations. These findings opens the door for using dynamical systems to analyze AI systems rather than the dominant paradigm of using AI to model dynamical systems.

Loading reusable skill documents into a bounded context window is now the primary way large language model (LLM) agents acquire task-specific capabilities, which makes skill selection a first-order determinant of task performance and token cost. Yet current agents score skills independently by semantic relevance and assemble the set by top-$k$ or greedy packing, with no quality guarantee or cost awareness on the selected set. As a result, redundant or poorly chosen skills waste scarce context tokens and can even degrade performance. We give the first model of how the selected skill set shapes execution outcomes and cast skill selection as an optimization problem: choose a skill set under a hard token budget to maximize a monotone submodular benefit minus context penalty. For this problem, we develop Best Prefix Selection (BPS), a polynomial-time algorithm, and prove, to our knowledge, the first performance guarantee for skill selection: a bicriteria $(1-1/e,1)$ approximation whose benefit coefficient is optimal in polynomial time. On a contamination-controlled BigCodeBench variant, BPS outperforms all the baselines, reaching $0.73$ measured task success versus $0.20$--$0.52$ for released skill routers, text retrievers, and the executor's own selection, on $28%$ fewer tokens than the strongest released router.

Direct Preference Optimization (DPO) has emerged as an effective approach for aligning large language models (LLMs) with human preferences. However, its adaptation to multimodal settings remains unexplored. Through representational analysis, we identify a key limitation in multimodal preference optimization, which we term visual insensitivity: models often fail to distinguish between images and those with critical visual context removed. Our theoretical analysis further uncovers two manifestations of this problem, namely Across-Image Insensitivity and Within-Image Insensitivity. To address these challenges, we propose Perception-Enhanced Alignment DPO (PEA-DPO), a framework for multimodal LLMs alignment, which explicitly leverages visual preference signals to overcome visual insensitivity. We further provide a theoretical analysis demonstrating that PEA-DPO provably mitigates both failure modes. Empirical results demonstrate that PEA-DPO enhances sensitivity to visual context while preserving the language modeling capacity of the base model. Evaluations across three hallucination benchmarks using MLLMs of varying scales show that PEA-DPO effectively mitigates visual insensitivity, achieves stronger multimodal alignment, and substantially reduces hallucinations.

Memory has become a key component of large language models, enabling them to retain information and learn from long-term interactions. However, existing memory benchmarks mainly evaluate whether information is correctly extracted, stored, and retrieved, while largely overlooking how retrieved memories reshape model reasoning and affect performance on the current task. We identify memory-induced cognitive traps: even faithfully recorded and semantically relevant memories can distort model reasoning or beliefs and degrade current task performance. To systematically evaluate these failure modes, we introduce MemTrapBench, which covers two forms of cognitive traps: Reasoning Fixation and Belief Distortion. Experiments across two model families and five representative memory frameworks show that MemTrapBench is challenging: all evaluated memory strategies underperform the no-memory setting, with even the strongest methods suffering drops of more than 10%. To mitigate these cognitive traps, we propose AdaptiveMem, a simple yet effective inference-time method that instructs LLMs to avoid memory traps. AdaptiveMem mitigates cognitive traps on MemTrapBench while preserving or improving performance on standard memory benchmarks across diverse memory frameworks.

Knowledge-Based Visual Question Answering (KB-VQA) relies on retrieving external information to answer queries involving long-tail entities. However, existing retrieval pipelines predominantly employ CLIP-style dual encoders, which prioritize surface-level visual similarity over entity-level semantic alignment. This paradigm often fails when semantically identical concepts exhibit large visual variations or when distinct entities appear visually similar. To address this, we propose KBMR, the first MLLM-based embedding retriever tailored for KB-VQA. Leveraging the robust autoregressive capabilities of MLLMs, KBMR maps images into a semantic space that better preserves concept identity. To tackle the challenge of noisy supervision in Wikipedia-scale retrieval, we introduce an MLLM-based semantic discriminator that generates continuous entity-consistency weights. These weights guide a novel continuous semantic distillation objective, enabling effective hard negative sampling and soft supervision beyond rigid binary labels. Extensive experiments demonstrate that KBMR significantly outperforms CLIP baselines, yielding up to a 14.7% improvement in retrieval Recall@1 and a 9.4% gain in end-to-end VQA accuracy. Code is available at https://github.com/realHarryX/KBMR.

Automated seizure detection from electroencephalography (EEG) is essential for continuous neurological monitoring, particularly for subclinical epileptic seizures that may exhibit only subtle electrographic changes. Existing time-series methods are often designed for fixed EEG channel configurations, thereby limiting their applicability to heterogeneous EEG recordings with irregular channel layouts. Although visual and language modeling offer promising alternatives, aligning heterogeneous EEG representations with clinical semantics remains challenging. We introduce ViTexSZ, a heterogeneous Vision-Text knowledge distillation framework for EEG seizure detection. ViTexSZ converts EEG recordings into structured waveform images and introduces a query-based multi-channel alignment module that maps source-dependent visual features into a unified token space. A heterogeneous teacher further integrates the aligned EEG representations with clinical prompts through a multimodal large language model, associating high-level clinical semantics with seizure-related evidence. Vision-text knowledge distillation then transfers the teacher representations to a lightweight student during detection. Experiments on four EEG seizure datasets demonstrate the generalizability of ViTexSZ across both subclinical and general seizure detection scenarios, achieving the highest accuracy on all datasets and relative improvements of up to 12.9% over the second-best baselines, showing its effectiveness.

We study the stability of minimal representations of controlled stochastic processes (in particular, transducers) under perturbations. This question is motivated by recent experiments finding predictive-state structure in the latent representations of neural networks. We consider standard, linear and predictive transducers. We introduce notions of approximate homomorphism capturing local structural similarity between them, together with metrics comparing their induced dynamics (which we refer to as interfaces), and prove properties such as composability of the approximate homomorphisms. For standard transducers, we show that there exist simple interfaces for which there is no approximate homomorphism between the different implementations of the dynamics. In contrast, for every finite-rank interface $\mathcal I$, we prove that all minimal linear transducers implementing interfaces sufficiently close to $\mathcal I$ have an approximate homomorphism to the minimal implementation of $\mathcal I$, with error linear in the perturbation size. We prove an analogous stability result for predictive transducers under a residual metric using some mild hypothesis regarding the indistinguishability of the belief states. These results identify conditions under which canonical transducer representations are robust to perturbations, while showing that such convergence fails without additional structural restrictions. Under the assumption that these type of abstractions are embedded into the hidden layers of modern AI models, this gives some theoretical support to the hypothesis that their latent representations exhibit structural convergence.

Dense causal attention remains expensive at long context even when implemented with highly optimized exact kernels. We study BF1, a deterministic block-aligned dyadic sparse-attention route that combines a small exact local neighborhood, a global first block, and logarithmically spaced historical blocks. The route is related to prior log-sparse and dilated attention patterns; our contribution is a correctness-gated pretrained-model retrofit, a matched topology-control study, and a systems characterization that connects per-layer sparsity to whole-model latency. For fixed block width, every converted layer uses O(n log n) selected token interactions and has O(log n) graph communication depth. On an NVIDIA RTX PRO 6000 Blackwell GPU, an optimized BF16 implementation crosses dense attention between 2K and 4K tokens and reaches a 10.91x per-layer prefill speedup at 32K. Retrofitting eight of 28 Qwen3-0.6B attention layers lowers warm whole-model time to first token by 7.7%, 11.3%, and 15.3% at 8K, 16K, and 32K, respectively, while the remaining dense layers keep the complete model asymptotically quadratic. Under a matched 1,000-step, 16.384M-token adaptation protocol, BF1 ranks first across three training seeds: mean report perplexity is 1.68639 versus 1.69154 for a matched static-random nonlocal graph, 1.69258 for dense continued training, and 1.81505 for equal-budget local sliding. At seed 1234, the packed-report paired interval places Dense-CT 0.3169-0.4055% above BF1 and static-random graph 17 0.2441-0.3642% above BF1. These results establish BF1 as a reproducible sparse operator and selective retrofit primitive with real long-context systems value. This paper evaluates numerical correctness, selected-interaction scaling, kernel performance, partial-model inference, and matched next-token language modeling.

Large language models (LLMs) are now embedded in scientific, educational, and governance workflows, with debates centering on their capabilities, mechanisms, and impacts. Yet these debates remain structured by persistent folk theories--intuitive, informal explanatory models that guide attitudes and actions. Deflationary slogans ("just autocomplete," "stochastic parrots," and "average of the internet") and anthropomorphic framings ("emergent agents" and "proto-minds") each capture genuine features of current systems but mistake those features for the whole. This Perspective proposes a minimal working model of LLM-based systems centered on four distinctions: between pretraining and deployed systems; between the learned distribution and particular samples; among parametric, contextual, and external memory; and between task competence and agency. The model is used to diagnose six misconceptions about LLMs: next-token prediction, regression to the mean, training-data regurgitation, model memory, alignment, and understanding. For each, the analysis identifies what the misconception gets right, which distinctions it conflates, and what follows for capability evaluation, system design, and governance. Applied to publisher AI policies as governance case studies, the framework shows both how policy language can conflate these distinctions and how such errors can be corrected. The model thereby avoids the parrot-mind binary by treating LLMs as simulators of discourse and task performance, offering a diagnostic toolkit for locating and correcting the errors these folk theories perpetuate.

On-policy distillation (OPD) has emerged as an effective framework for post-training language models by pairing student-generated trajectories with dense token-level supervision from a teacher. However, OPD implicitly assumes that teacher-derived rewards are an appropriate proxy for reasoning progress, and therefore treats all teacher feedback equally during policy optimization. While in practice, this assumption does not always hold. We observe that teacher-derived rewards often conflict with genuine reasoning progress, as reasoning steps with clear reasoning advancement may still receive lower distillation rewards, simply due to deviation from teacher's outputs. To address this mismatch, we propose Reasoning-Progress-Aware Reward Filtering for On-Policy Distillation (R2-OPD), which constructs two within-trajectory rankings of reasoning spans, one from teacher-derived rewards and the other from independently estimated progress reward. Distillation rewards are selectively suppressed whenever the two rankings disagree, reducing supervision that conflicts with reasoning progress while preserving effective teacher guidance. Our approach shows consistent improvement over standard OPD especially regarding reasoning performances.

We introduce Deep-MKV-TS, a path-dependent McKean-Vlasov framework for financial scenario generation. The stochastic dynamics are chosen by matching selected path and volatility features of generated scenarios to those observed in the data. Starting from an interpretable reference model, Deep-MKV-TS preserves the reference drift and adjusts its volatility, while a regularization penalty limits unnecessary departures from the calibrated dynamics. We solve the resulting control problem using a neural, sample-based implementation of the stochastic maximum principle. We validate the method against an exactly computable oracle. On Heston and Heston-mixture models, Deep-MKV-TS substantially reduces path-dependent and volatility-related deficiencies of the reference model. In delayed-volatility experiments, the correction remains effective as the forecasting horizon increases, while direct training becomes less reliable. On held-out intraday equity-index futures, the corrected model improves conditional forecasts relative to the reference and reaches a level of performance comparable to flexible generative and historical baselines. The resulting scenarios also support greater exposure than the reference under a fixed drawdown-risk target. These results show that path-dependent McKean-Vlasov control can enrich an interpretable reference model without replacing it.

World models increasingly treat compact multimodal representations as interfaces between perception and physical interaction, yet existing probes do not establish whether different sensors carry the same executable meaning or whether that meaning survives a new action composition. We introduce an operational capability hierarchy and the Disjoint-Bridge Operator-Substitution Certificate (DBOSC), which asks whether independently trained modality compilers enter a frozen response chart interchangeably on evidence outside their training panels. On Cluster Haptic, audio and acceleration representations of the same unseen surface are 4.5x closer in response space than wrong-surface pairings, with the gap holding for all 19 held-out surfaces; unsealing withheld responses confirms that every branch predicts the physics better than the population chart. We then test ordered execution in a controlled elastoplastic system with complementary modality blind spots. At the pre-registered budget, the prerequisite refuses the stack because the frozen executor cannot advance even an exact chart coordinate through a held-out program. At a converged budget, the same rank-three chart executes those programs (oracle NMSE 0.18), fusion improves on both modalities, and 14 of 16 registered checks pass; the two failures arise because a diagonal restriction of the fused information matrix performs as well as the full one. Clearing the gate is a property of the executor, not the chart: an executor emitting whole programs instead of shared per-step dynamics is 38x worse than an entity-blind predictor on the same chart. A matching non-identifiability result explains why compression and fusion alone cannot determine an unseen composition law. These results separate attribute access, response substitution, fusion closure, and ordered execution into distinct, separately testable achievements.