127,664Total entries

60Added by human

127,604Found by agent

13Categories

Knowledge catalogue

Search: “reasoning”

19,026 results

Safety

On-Policy Attention Linearization

DGX agent

arXiv:2609.31947v1 Announce Type: new Abstract: Hybrid transformer architectures that replace most softmax attention layers with linear attention offer transformer-level quality at a fraction of the m

safetyarxiv-cs-lg

29 Sept 2026

Research

On Temporal Binding in Large Audio Language Models

DGX agent

arXiv:2609.34806v1 Announce Type: cross Abstract: Reasoning about temporal structure of audio recordings requires Large Audio Language Models (LALMs) to associate sound events with their temporal posi

Search: “reasoning”

On-Policy Attention Linearization

On Temporal Binding in Large Audio Language Models

ReSight-SMC: Two-Stage Power Sampling via Island SMC with Visual Scouts

SCOPD: Sparse-Context On-Policy Self-Distillation for Efficient Vision-Language Models

Selecting Diverse SFT Traces Improves Post-RL Generalization

Do LLMs Understand Context? A Knowledge Graph-Based Evaluation Framework

Evolutionary Safety of Recursive Self-Improving AI: Taxonomy, Risk Discovery, and Evaluation

Governed Deduction: Policy-Grounded Premise Authorization Beyond Relevance

Highlight-Then-Summarize: Learning to Compress Evidence for Long-Context Understanding

When Is a Multi-Agent Code Judge Actually Grounded? Two Label-Free Measurements, and a Judge That Declines to Guess

GPT-6-Astra Lights Up Embodied Navigation: Evaluation in Zero-Shot Vision-and-Language Navigation in Continuous Environments

HarnessPAI: An Evolving Harness for Physical AI

DCRL: Decoupling and Coupling Reinforcement Learning via Policy-Reward Manifold Alignment

Do We Need Complex Topology Control? Distinct-Peer Random Routing Improves Cost-Efficiency in Sparse Multi-Agent Debate

Energy-guided Recursive Model

ReasonLab: A Controlled and Auditable Evaluation of Prompting Techniques for Multiple-Choice QA

Collapse, Not Complexity: Failure-Conditioned Decomposition Repair for End-to-End Document Parsing

CTRL: Control-Based Time Series Forecasting with LLM-Guided Residual Learning

Information-Time Proximal Policy Optimization

Tool-Augmented On-Policy Distillation for LLM Domain Adaptation in Sequence-Based Omics Tasks

Beyond Exact Match: Task-Aware GRPO for Cross-Domain PCBA Visual Question Answering

PRIME: Perception Feedback with Situational Memory Embeddings in VLA Models

Attention Dispersion as a Diagnostic Signal for Hallucination in Large Language Models

CapMem: A Benchmark for Caption-Based Episodic Memory in Egocentric Video

HALT: Hallucination Assessment via Log-probs as Time series

Label-free steering: Compressing test-time reinforcement learning into bias-only subspaces

MUSE: Benchmarking Large Vision-Language Models on Multi-Modal Understanding in Situated Education

UAVs Meet Embodied Intelligence: Bridging Human Intents and Flying Dynamics Via Harnessing Physical-Digital AI Agents

Visual Compliance via Executable Safety Rule Entailment

Available but Unclaimed: An Empirical Study of Human-AI Synergy

Script Fragmentation and Format: What Drives the English-Bengali Performance Gap in Open LLMs?

Beyond Depth and Width: The Information-Slack Dilemma in Streaming Test-Time Compute

Beyond OCR Accuracy: Text-Centric VQA Under Image Degradation with Modular and End-to-End

Bridging Thought and Action: Taming Long-Horizon Instability in Open-Source LLM Agents with a MetaTool-Enhanced ROS Framework

Compositional SVG Generation via VLM-Driven Hierarchical Semantic Parsing

HISPO: Hierarchical Importance-Sampling Policy Optimization with Entropy-Derived Segments

MedVA: An End-to-End Neuro-Symbolic Agentic System for Medical Volume Visualization

SALUTE: Benchmarking and Adapting LLMs for the Defense Domain

R2VC: Modular Fact-Checking with Retrieval, Verification, and Confidence Calibration

Caption-once, Frames-on-Demand: Visual-Need Routing for Budget-Aware Agentic Long Video Understanding

LexAgentHallu: A Hierarchical Benchmark for Profiling Hallucinations in Legal Agents

OpenResearcher: A Fully Open Pipeline for Long-Horizon Deep Research Trajectory Synthesis

Towards Reliable Medical LLMs: Benchmarking and Enhancing Confidence Estimation of Large Language Models in Medical Consultation

From Version Conflicts to Decision Conflicts: Selective Revalidation for Long-Running AI Agents

SkillAlign: Aligning Skill Interfaces for LLM-based Agents

What Does MMLU Actually Measure? A Psychometric Audit of Difficulty Structure in Aggregate Benchmark Scores

3DHarnessBench: Probing Agentic 3D-to-Code Capabilities of Frontier Vision-Language Models

BanglaMemeX: Advancing Cultural Metaphoric Image Interpretation in Bangla with a Multimodal Explainable Dataset