Search: “reasoning”

SURGENT: A Surgical Multi-Agent Assistance System Across the Perioperative Workflow

arXiv:2605.29368v1 Announce Type: cross Abstract: The intricate nature of modern surgical care necessitates intelligent systems that can synthesize extensive patient records, support collaborative dec

29 May 2026

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Research

Token Inflation: How Dishonest Providers Can Overcharge for Large Language Model Usage

DGX agent

arXiv:2605.30040v1 Announce Type: cross Abstract: Per-token billing is now the standard pricing model for commercial large language models (LLMs), so the honesty of reported token counts directly affe

researcharxiv-cs-ai

29 May 2026

Local Ai

Unifying Temporal and Structural Credit Assignment in LLM-Based Multi-Agent Prompt Optimization

arXiv:2605.30227v1 Announce Type: cross Abstract: While Multi-Agent Systems (MAS) empower Large Language Models to tackle complex reasoning tasks through collaborative interaction, optimizing their dy

local-aiarxiv-cs-ai

29 May 2026

Xetrieval: Mechanistically Explaining Dense Retrieval

arXiv:2605.29507v1 Announce Type: new Abstract: Explaining why dense retrievers assign high relevance scores remains challenging because retrieval decisions are made through opaque high-dimensional em

29 May 2026

ADWIN: Adaptive Windows for Horizon-Aware On-Policy Distillation

arXiv:2605.28396v1 Announce Type: cross Abstract: On-policy distillation (OPD) transfers reasoning behavior by training a student on teacher feedback along student-generated trajectories, but standard

safetyarxiv-cs-ai

Beyond Binary Moral Judgment: Modeling Ethical Pluralism in AI

arXiv:2605.28707v1 Announce Type: new Abstract: Critical decision-making in socially consequential spaces is increasingly involving AI systems at varying capacities. Yet, despite the ubiquity of auton

CPPO: Contrastive Perception Policy Optimization for VLM Agents

arXiv:2601.00501v2 Announce Type: replace Abstract: We introduce CPPO, a Contrastive Perception Policy Optimization method for finetuning vision--language models (VLMs). Reliable perception is a core

safetyarxiv-cs-cv

Cyclical Entropy Eruption: Entropy Dynamics in Agent Reinforcement Learning

arXiv:2605.27954v1 Announce Type: new Abstract: Agentic large language models are increasingly used to solve real-world tasks by reasoning over goals, invoking tools, and interacting with external env

agentsarxiv-cs-lg

Tutorials

Human Label Variation as Stable Signal: Learning Annotator-Specific Explanation Behavior via Cross-Annotator Preference Optimization

arXiv:2605.28802v1 Announce Type: new Abstract: Free-text explanations extend human label variation (HLV) beyond label disagreement by revealing the reasoning and preferences behind annotators' decisi

tutorialsarxiv-cs-cl

IRDS: Interpretable RLVR Data Selection via Verifier-Coupled Sparse Autoencoder Coverage

arXiv:2605.28247v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards (RLVR) has become a key technique for en- hancing LLM reasoning, yet its data ineffi- ciency remains a

Joint Training of Multi-Token Prediction in Reinforcement Learning via Optimal Coefficient Calibration

arXiv:2605.28184v1 Announce Type: new Abstract: Reinforcement Learning from Verifiable Rewards (RLVR) has emerged as the standard paradigm for improving reasoning capability of large language models,

safetyarxiv-cs-lg

Long Live The Balance: Information Bottleneck Driven Tree-based Policy Optimization

arXiv:2605.28109v1 Announce Type: new Abstract: Recent advances in online reinforcement learning (RL) for large language models (LLMs) have demonstrated promising performance in complex reasoning task

safetyarxiv-cs-lg

MemCog: From Memory-as-Tool to Memory-as-Cognition in Conversational Agents

arXiv:2605.28046v1 Announce Type: new Abstract: Existing agent memory systems universally follow what we term a Memory-as-Tool paradigm where a single query triggers one-shot retrieval of flat passage

MemGuard: Preventing Memory Contamination in Long-Term Memory-Augmented Large Language Models

arXiv:2605.28009v1 Announce Type: cross Abstract: Memory-augmented large language models extend reasoning beyond a fixed context window by maintaining long-term memory across interactions. However, ex

Playing with Words, Improving with Rewards: Training Language Models for Creative Association

arXiv:2605.27832v1 Announce Type: new Abstract: Large Language Models (LLMs) are being applied to increasingly difficult problems and use cases. To navigate their vast solution spaces effectively, LLM

researcharxiv-cs-cl

ResearchMath-14K: Scaling Research-Level Mathematics via Agents

arXiv:2605.28003v1 Announce Type: new Abstract: The frontier of mathematics is defined by problems whose solutions are not yet known, yet it remains unclear whether language models can meaningfully en

agentsarxiv-cs-cl

Trust Me, I'm an Expert: Decoding and Steering Authority Bias in Large Language Models

arXiv:2601.13433v3 Announce Type: replace Abstract: Prior research demonstrates that performance of language models on reasoning tasks can be influenced by suggestions, hints and endorsements. However

safetyarxiv-cs-cl

Understanding Automated Program Repair Agents Through the Lens of Traceability: An Empirical Study

arXiv:2506.08311v2 Announce Type: replace-cross Abstract: Automated Program Repair (APR) agents leverage Large Language Models (LLMs) to autonomously diagnose and fix software bugs through reasoning,

UserHarness: Harnessing User Minds for Stronger Agent Theory-of-Mind

arXiv:2605.27721v1 Announce Type: cross Abstract: Understanding what a user believes and intends is central to building effective agent assistants. This ability is often evaluated through Theory-of-Mi

Advancing Creative Physical Intelligence in Large Multimodal Models

arXiv:2605.26396v1 Announce Type: new Abstract: Large multimodal models (LMMs) have rapidly advanced in perception and reasoning; however, it remains unclear whether these capabilities generalize to d

Beyond a Single Direction: Chain-of-Thought Disrupts Simple Steering of Refusal

arXiv:2605.26772v1 Announce Type: new Abstract: Large reasoning models (LRMs) generate chain-of-thought (CoT) traces before producing final outputs, introducing a dynamic internal state that may compl

Enabling Extensible Embodied Capabilities with Tools

arXiv:2605.26637v1 Announce Type: new Abstract: Most existing embodied intelligence methods formulate perception, reasoning, planning, and control within a unified parameterized policy. Yet these capa

safetyarxiv-cs-ro

InfoSynth: Information-Guided Benchmark Synthesis for LLMs

model-releasesarxiv-cs-cl

arXiv:2601.00575v2 Announce Type: replace Abstract: Large language models (LLMs) have demonstrated significant advancements in reasoning and code generation, but efficiently creating new benchmarks to

Learning to Act under Noise: Enhancing Agent Robustness via Noisy Environments

arXiv:2605.27209v1 Announce Type: new Abstract: Recent advances in large language models (LLMs) have facilitated the widespread deployment of LLMs as interactive agents capable of reasoning, planning,

Memory Architectures for Multi-Turn Text-to-SQL: A Benchmark and Empirical Study

model-releasesarxiv-cs-cl

arXiv:2605.26394v1 Announce Type: new Abstract: Multi-turn Text-to-SQL is central to enterprise analytics yet remains predominantly evaluated in single-turn settings. We introduce EnterpriseMem-Bench,

QUACK: Questioning, Understanding, and Auditing Communicated Knowledge in Multimodal Social Deduction Agents

arXiv:2605.27068v1 Announce Type: cross Abstract: Social deduction games have become a popular testbed for probing reasoning, deception, coordination, and belief modeling in Large Language Model (LLM)

TPS-Drive: Task-Guided Representation Purification for VLM-based Autonomous Driving

arXiv:2605.27038v1 Announce Type: new Abstract: Vision-Language Models (VLMs) provide a promising foundation for autonomous driving planning, yet bridging semantic reasoning and precise 3D spatial for

safetyarxiv-cs-ro

Tutorials

Transformers with RL or SFT Provably Learn Sparse Boolean Functions, But Differently

arXiv:2511.17852v2 Announce Type: replace Abstract: Transformers can acquire Chain-of-Thought (CoT) capabilities to solve complex reasoning tasks through fine-tuning. Reinforcement learning (RL) and s

tutorialsarxiv-cs-lg

AION: Next-Generation Tasks and Practical Harness for Time Series

arXiv:2605.25045v1 Announce Type: new Abstract: Time series research is moving beyond fixed forecasting benchmarks toward realistic tasks that combine prediction, contextual reasoning, tool use, and s

All Leaks Count, Some Count More: Interpretable Temporal Contamination Detection and Mitigation in LLM Backtesting

arXiv:2602.17234v2 Announce Type: replace Abstract: Backtesting LLMs on resolved events assumes models reason only from pre-cutoff knowledge, yet pretrained models inevitably leak post-cutoff knowledg

Asking LLMs to Verify First is Almost Free Lunch

arXiv:2511.21734v2 Announce Type: replace-cross Abstract: To enhance the reasoning capabilities of Large Language Models (LLMs) without high costs of training, nor extensive test-time sampling, we int

Can LLMs Time Travel? Enhancing Temporal Consistency in Legal Agentic Search through Reinforcement Learning

arXiv:2605.25920v1 Announce Type: cross Abstract: While large language models (LLMs) augmented with agentic search capabilities show promise for legal reasoning, they overlook a fundamental constraint

CausalFlow: Causal Attribution and Counterfactual Repair for LLM Agent Failures

arXiv:2605.25338v1 Announce Type: cross Abstract: Large language model (LLM) agents frequently fail on multi-step tasks involving reasoning, tool use, and environment interaction. While such failures

Applications

ConceptM^3oE: Concept-Guided Multimodal Mixture of Experts for Interpretable Computational Pathology

arXiv:2605.24399v1 Announce Type: new Abstract: Healthcare models are transitioning from unimodal prediction toward multimodal reasoning over heterogeneous diagnostic inputs. In computational patholog

applicationsarxiv-cs-ai

Contextual Rollout Bandits for Reinforcement Learning with Verifiable Rewards

arXiv:2602.08499v2 Announce Type: replace-cross Abstract: Reinforcement Learning with Verifiable Rewards (RLVR) is an effective paradigm for improving the reasoning capabilities of large language mode

safetyarxiv-cs-ai

Discounted Beta-Bernoulli Reward Estimation for Sample-Efficient Reinforcement Learning with Verifiable Rewards

arXiv:2603.18444v2 Announce Type: replace-cross Abstract: Reinforcement learning with verifiable rewards (RLVR) has emerged as an effective post-training paradigm for improving the reasoning capabilit

safetyarxiv-cs-ai

DiscoverPhysics: Benchmarking LLMs for Out-of-the-Box Scientific Thinking

model-releasesarxiv-cs-lg

arXiv:2605.26087v1 Announce Type: cross Abstract: Frontier LLMs now perform strongly across a wide range of physics evaluations, but it is hard to disentangle genuine reasoning from recall of establis

Applications

Language Movement Primitives: Grounding Language Models in Robot Motion

arXiv:2602.02839v3 Announce Type: replace Abstract: Enabling robots to perform novel manipulation tasks from natural language instructions remains a fundamental challenge in robotics, despite signific

applicationsarxiv-cs-ro

PathMem: Toward Cognition-Aligned Memory Transformation for Pathology MLLMs

arXiv:2603.09943v2 Announce Type: replace Abstract: Computational pathology demands both visual pattern recognition and dynamic integration of structured domain knowledge, including taxonomy, grading

RL with Learnable Textual Feedback: A Bilevel Approach

model-releasesarxiv-cs-lg

arXiv:2605.24547v1 Announce Type: new Abstract: Reinforcement learning with verifiable rewards can improve LLM reasoning, but learning remains sample-inefficient when terminal rewards are sparse. This

Scaling Natural-Language Graph-Based Test Time Compute for Automated Theorem Proving

arXiv:2503.11657v3 Announce Type: replace Abstract: Large language models have demonstrated remarkable capabilities in natural language processing tasks requiring multi-step logical reasoning capabili

researcharxiv-cs-cl

Test-Time Deep Thinking to Explore Implicit Rules

arXiv:2605.24828v1 Announce Type: new Abstract: With the continuous advancement of Large Language Models (LLMs), intelligent agents are becoming increasingly vital. However, these agents often fail in

DRIVESPATIAL: A Benchmark for Spatiotemporal Intelligence in VLMs for Autonomous Driving

model-releasesarxiv-cs-cv

arXiv:2605.23176v1 Announce Type: new Abstract: Spatiotemporal intelligence in autonomous driving (AD) requires an agent to integrate multi-view observations into a coherent scene representation, main

General Hazard Detection

arXiv:2605.23304v1 Announce Type: new Abstract: Hazard, as an abstract concept, is typically defined through cognitive-level logical reasoning rather than concrete examples. In contrast, existing haza

safetyarxiv-cs-cv

Tutorials

Learnability-Informed Fine-Tuning of Diffusion Language Models

arXiv:2605.22939v1 Announce Type: new Abstract: We aim to improve the reasoning capabilities of diffusion language models (DLMs). While SFT is a popular post-training recipe for autoregressive models,

tutorialsarxiv-cs-cl

R^3L: Reflect-then-Retry Reinforcement Learning with Language-Guided Exploration, Pivotal Credit, and Positive Amplification

arXiv:2601.03715v2 Announce Type: replace-cross Abstract: Reinforcement learning drives recent advances in LLM reasoning and agentic capabilities, yet current approaches struggle with both exploration

SPACENUM: Revisiting Spatial Numerical Understanding in VLMs

arXiv:2605.23898v1 Announce Type: new Abstract: Vision-Language Models (VLMs) are increasingly deployed in embodied environments, where they need produce numerical outputs such as action magnitudes an

What Training Data Teaches RL Memory Agents: An Empirical Study of Curriculum Effects in Memory-Augmented QA

model-releasesarxiv-cs-cl

arXiv:2605.23067v1 Announce Type: new Abstract: Reinforcement learning (RL) has emerged as a viable recipe for training LLM agents to reason over external memory banks in multi-session dialogue. Exist