arXiv:2509.21275v3 Announce Type: replace-cross Abstract: Long context training is crucial for LLM’s context extension. Existing schemes, such as sequence parallelism, incur substantial communication overhead. Pipeline parallelism (PP) reduces this cost, but its effectiveness hinges on partitioning granularity. Batch-level PP employing sequence packing exhibits high memory consumption in long-context scenarios, whereas token-level PP splitting sequences into […]
MM-JudgeBias: A Benchmark for Evaluating Compositional Biases in MLLM-as-a-Judge
arXiv:2604.18164v3 Announce Type: replace-cross Abstract: Multimodal Large Language Models (MLLMs) have been increasingly used as automatic evaluators-a paradigm known as MLLM-as-a-Judge. However, their reliability and vulnerabilities to biases remain underexplored. We find that many MLLM judges fail to reliably integrate key visual or textual cues, yielding unreliable evaluations when evidence is missing or mismatched, and […]
Words that make SENSE: Sensorimotor Norms in Learned Lexical Token Representations
arXiv:2602.00469v2 Announce Type: replace-cross Abstract: While word embeddings derive meaning from co-occurrence patterns, human language understanding is grounded in sensory and motor experience. We present $textSENSE$ $(textbfStextensorimotor $ $textbfEtextmbedding $ $textbfNtextorm $ $textbfStextcoring $ $textbfEtextngine)$, a learned projection model that predicts Lancaster sensorimotor norms from word lexical embeddings. We also conducted a behavioral study where […]
Supervised Learning Has a Necessary Geometric Blind Spot: Theory, Consequences, and Minimal Repair
arXiv:2604.21395v1 Announce Type: cross Abstract: We prove that empirical risk minimisation (ERM) imposes a necessary geometric constraint on learned representations: any encoder that minimises supervised loss must retain non-zero Jacobian sensitivity in directions that are label-correlated in training data but nuisance at test time. This is not a contingent failure of current methods; it is […]
Model Capability Assessment and Safeguards for Biological Weaponization
arXiv:2604.19811v2 Announce Type: replace-cross Abstract: AI leaders and safety reports increasingly warn that advances in model reasoning may enable biological misuse, including by low-expertise users, while major labs describe safeguards as expanding but still evolving rather than settled. This study benchmarks ChatGPT 5.2 Auto, Gemini 3 Pro Thinking, Claude Opus 4.5 and Meta’s Muse Spark […]
Reasoning on the Manifold: Bidirectional Consistency for Self-Verification in Diffusion Language Models
arXiv:2604.16565v2 Announce Type: replace-cross Abstract: While Diffusion Large Language Models (dLLMs) offer structural advantages for global planning, efficiently verifying that they arrive at correct answers via valid reasoning traces remains a critical challenge. In this work, we propose a geometric perspective: Reasoning on the Manifold. We hypothesize that valid generation trajectories reside as stable attractors […]
On the Role of Preprocessing and Memristor Dynamics in Reservoir Computing for Image Classification
arXiv:2604.21602v1 Announce Type: cross Abstract: Reservoir computing (RC) is an emerging recurrent neural network architecture that has attracted growing attention for its low training cost and modest hardware requirements. Memristor-based circuits are particularly promising for RC, as their intrinsic dynamics can reduce network size and parameter overhead in tasks such as time-series prediction and image […]
From Noise to Intent: Anchoring Generative VLA Policies with Residual Bridges
arXiv:2604.21391v1 Announce Type: cross Abstract: Bridging high-level semantic understanding with low-level physical control remains a persistent challenge in embodied intelligence, stemming from the fundamental spatiotemporal scale mismatch between cognition and action. Existing generative VLA policies typically adopt a “Generation-from-Noise” paradigm, which disregards this disparity, leading to representation inefficiency and weak condition alignment during optimization. In […]
StructMem: Structured Memory for Long-Horizon Behavior in LLMs
arXiv:2604.21748v1 Announce Type: cross Abstract: Long-term conversational agents need memory systems that capture relationships between events, not merely isolated facts, to support temporal reasoning and multi-hop question answering. Current approaches face a fundamental trade-off: flat memory is efficient but fails to model relational structure, while graph-based memory enables structured reasoning at the cost of expensive […]
Quantifying how AI Panels improve precision
arXiv:2604.16432v2 Announce Type: replace-cross Abstract: AI in applications like screening job applicants had become widespread, and may contribute to unemployment especially among the young. Biases in the AIs may become baked into the job selection process, but even in their absence, reliance on a single AI is problematic. In this paper we derive a simple […]
Directional Confusions Reveal Divergent Inductive Biases Through Rate-Distortion Geometry in Human and Machine Vision
arXiv:2604.21909v1 Announce Type: cross Abstract: Humans and modern vision models can reach similar classification accuracy while making systematically different kinds of mistakes – differing not in how often they err, but in who gets mistaken for whom, and in which direction. We show that these directional confusions reveal distinct inductive biases that are invisible to […]
Conjecture and Inquiry: Quantifying Software Performance Requirements via Interactive Retrieval-Augmented Preference Elicitation
arXiv:2604.21380v1 Announce Type: cross Abstract: Since software performance requirements are documented in natural language, quantifying them into mathematical forms is essential for software engineering. Yet, the vagueness in performance requirements and uncertainty of human cognition have caused highly uncertain ambiguity in the interpretations, rendering their automated quantification an unaddressed and challenging problem. In this paper, […]