Search: “reasoning”

Training LLM Judges from Language Feedback via Position-Selective Self-Distillation

arXiv:2609.38792v1 Announce Type: new Abstract: We study training LLM judges from natural language feedback, especially for subjective tasks where the verdict depends strongly on which evaluation crit

researcharxiv-cs-cl

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Tutorials

TripleFlow: Training-Free Video Object Removal by Bridging Residual Editing and Native Generation

DGX agent

arXiv:2609.39157v1 Announce Type: new Abstract: Video object removal presents a uniquely difficult editing challenge. Because a removal prompt specifies only what to erase rather than what to generate

tutorialsarxiv-cs-cv

1 Oct 2026

Local Ai

Trust the Critic More

arXiv:2609.39247v1 Announce Type: cross Abstract: Standard language model RL algorithms credit every token of a long rollout with the same advantage determined by the terminal reward. Actor-critic met

local-aiarxiv-cs-ai

Trustworthy Runtime Error Healing in Real-World Repositories: A Benchmark and Guardrail

arXiv:2609.39086v1 Announce Type: cross Abstract: Runtime error healing lets a crashed program continue by generating code that repairs its live runtime state. Recent work shows that LLMs can generate

Vision-Language-Action Autonomous Driving Agent with Language-based Memory

arXiv:2609.38641v1 Announce Type: new Abstract: Vision-Language-Action (VLA) foundation models have recently emerged as one of the prevailing solutions for autonomous driving, as they can utilize know

agentsarxiv-cs-cv

WARP: A Unified Benchmark for Invisible Image Watermarking -- Robustness and Protection Against Attacks

arXiv:2609.40031v1 Announce Type: cross Abstract: Digital image watermarking is increasingly critical in media contexts, as emerging regulations and industry practices require marking AI-generated con

When Does a Spoken Agent Have Enough Evidence to Act? The PACT-SLM Contract Test

arXiv:2609.38232v1 Announce Type: cross Abstract: Streaming spoken agents may take an external action before the available speech supports it, yet final-turn scores do not reveal whether each observed

agentsarxiv-cs-cl

When Does Randomized Oversight Align AI Agents That Can Conceal?

arXiv:2609.38262v1 Announce Type: cross Abstract: Oversight changes the evidence it relies on. We ask when randomized audits and scoring align AI agents that can conceal misconduct and alter records.

agentsarxiv-cs-ai

Why am I like this? (Full Chat and Image generation on a 286 Tandy 1000 TL/3)

model-releasesr-localllama

40 year tech gap? No problem! The Tandy runs DeskMind, a native DOS program. It talks over WiFi (a PicoMEM 2 card with mTCP) to a small Python server on my PC. That server drives Qwen3.8-27B (NInfer o

X-Planner: Event-Structured Task Planning for Embodied Intelligence

arXiv:2609.25187v2 Announce Type: replace Abstract: Task planning bridges high-level instructions and executable behavior in long-horizon manipulation, yet modern Vision-Language-Action (VLA) systems

researcharxiv-cs-ai

ABC: Advantage-Based Control Variates for Reinforcement Learning with Verifiable Rewards

arXiv:2609.36058v1 Announce Type: new Abstract: Recent progress in reinforcement learning with verifiable rewards (RLVR) has highlighted the effectiveness of simple critic-free policy-gradient methods

AdvancedMathBench: A Benchmark Suite for Advanced Mathematical Proof Generation and Verification

model-releasesarxiv-cs-cl

arXiv:2607.11849v2 Announce Type: replace Abstract: Large language models (LLMs) have achieved remarkable performance on high-school and competition-level mathematics, yet their capabilities on advanc

Agentic Commerce Bench: Measuring Fraud Detection for Agents That Spend Money

arXiv:2609.35886v1 Announce Type: cross Abstract: AI agents now hold spend authority and settle payments without per-action human confirmation. The resulting loss is often not a security failure: a co

Auditable Long-Term Memory: A Deterministic Retrieval Chain Measured at 479/475 of 500 on LongMemEval-S

arXiv:2609.38021v1 Announce Type: cross Abstract: We evaluate an auditable long-term memory system on LongMemEval-S. Its retrieval chain uses hybrid candidate retrieval, cross-encoder reranking, cover

Beam Search as Test-Time Self-Distillation via Counterfactual Contexts

arXiv:2609.37041v1 Announce Type: cross Abstract: Self-Distillation Fine-Tuning (SDFT) enables a language model to act as its own teacher: by conditioning on a demonstration, the model produces an imp

Beyond Low-Rank Parameterization: Narrowing the Gap Between LoRA and Full Fine-Tuning via Gradient Decomposition

arXiv:2609.37027v1 Announce Type: new Abstract: Low-Rank Adaptation (LoRA) is a widely used approach to parameter-efficient fine-tuning (PEFT), yet a performance gap can remain relative to full fine-t

BRIDGE: Bilevel Retrieval-Credit-Aware Agentic Reinforcement Learning

arXiv:2609.36505v1 Announce Type: new Abstract: Agentic reinforcement learning (ARL) with verifiable rewards improves the ability of large language models (LLMs) to tackle knowledge-intensive tasks by

Code4Scene: Benchmarking Coding Agents for Constructing and Editing 3D Scenes

arXiv:2609.36777v1 Announce Type: new Abstract: Frontier coding agents can now write and execute code that authors 3D environments, but whether they reliably understand 3D structure and precisely cont

Concealing LLM-Based Multi-Agent Topology via Phantom Structure Injection

arXiv:2609.37567v1 Announce Type: cross Abstract: Driven by the rapid advancement of large language models (LLMs), LLM-based multi-agent systems (MAS) have emerged as a powerful paradigm for collabora

Dating the Model: Hidden Dates in System Prompts Affect LLM Evaluation

arXiv:2609.36931v1 Announce Type: new Abstract: Reproducibility is essential for scientific research, yet prior work shows that LLM outputs vary with hardware and batching. We identify an overlooked f

researcharxiv-cs-cl

Dynamic Optimizations of LLM Ensembles with Two-Stage Reinforcement Learning Agents

arXiv:2502.04492v3 Announce Type: replace Abstract: The advancement of LLMs and their accessibility have triggered renewed interest in multi-agent reinforcement learning as robust and adaptive framewo

safetyarxiv-cs-cl

Emergent phases of superposition: from partial to full representation

arXiv:2609.36455v1 Announce Type: new Abstract: Large language models are thought to represent features by vectors in a hidden space of dimension given by the model's width. Superposition, in which mo

researcharxiv-cs-lg

Explore, Execute, Evolve: A Skill Acquisition and Reuse Loop for Embodied Agents

arXiv:2609.37810v1 Announce Type: cross Abstract: Vision-language-action and world-action models have demonstrated impressive capabilities in robotics, yet generalization to unseen tasks remains chall

agentsarxiv-cs-ai

Gondola: Grounded Vision Language Planning for Robotic Manipulation

arXiv:2506.11261v2 Announce Type: replace-cross Abstract: Vision-language-action (VLA) models have shown promising progress in robotic manipulation. However, directly mapping visual observations and l

Industry

Half a million downloads in a month. Today, our open source family takes another step forward. Thank you for the incredible support behind o…

Half a million downloads in a month. Today, our open source family takes another step forward. Thank you for the incredible support behind our first-generation models. We’re excited to introduce TwIL-

industryclem-delangue--x

Interactive-Policy Distillation with Bidirectional Propose-and-Verify

arXiv:2609.36546v1 Announce Type: cross Abstract: On-policy distillation (OPD) trains a student model on its self-generated trajectories with dense token-level teacher feedback. However, naive OPD may

Interpolated Policy Distillation: A Controllable Continuum Between Off-Policy and On-Policy Distillation

arXiv:2609.37170v1 Announce Type: cross Abstract: Off-policy and on-policy distillation have traditionally been formulated as separate paradigms, each favoring a different property of distillation tra

Is Human-Readable Text Necessary for Effective LLM Fine-Tuning?

arXiv:2609.35868v1 Announce Type: new Abstract: Is human readability necessary for effective fine-tuning of large language models? We investigate whether model-conditioned training representations can

It's All Training: A Fully Synthetic Single-Stage Recipe for LLMs

arXiv:2609.37891v1 Announce Type: new Abstract: Current pre-training datasets are derived from web crawls, with all their issues, and were not designed to support mid- and post-training pipelines--for

researcharxiv-cs-cl

Learning from Teacher Continuations at Student States

model-releasesarxiv-cs-cl

arXiv:2609.36246v1 Announce Type: new Abstract: We present OLIVE (OnLine InterVEntion). At each iteration, the evolving student policy generates a new prefix, the teacher continues it autoregressively

Learning from the Gap Between Pass@K and Pass@1

model-releasesarxiv-cs-lg

arXiv:2609.35793v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly trained with reinforcement learning from verifiable rewards (RLVR). An exact verifier can also support tes

Mixture of Self-Improving Branches For Agent Harness Optimization

arXiv:2609.37834v1 Announce Type: new Abstract: Harness optimization provides a practical setting for recursive self-improvement (RSI), where agent-generated modifications inform subsequent changes th

Pointwise or Pairwise: When Do Pairwise Losses Help Reward Learning, Provably?

arXiv:2609.37209v1 Announce Type: new Abstract: Pairwise losses are increasingly used for reward learning even when pointwise rewards are observed, with mixed empirical results. When and why do pairwi

PolyOCR-Venus: Unified OCR Foundation Models for Text-Centric Visual Intelligence

arXiv:2609.37712v1 Announce Type: cross Abstract: Optical Character Recognition (OCR) is evolving from plain-text transcription toward general visual intelligence, requiring models to recognize, local

PR-OPD: Privileged Representation On-policy Self-Distillation for Agentic Reinforcement Learning

arXiv:2609.36642v1 Announce Type: new Abstract: Language-model agents are usually trained by reinforcement learning from one reward per episode, and privileged self-distillation enriches it by letting

Prompts Live on an Arc: Gaussian Curricula in Fisher--Rao Coordinates for Rollout-Efficient GRPO

arXiv:2609.38018v1 Announce Type: new Abstract: Group relative policy optimization (GRPO) learns only from prompts whose sampled responses disagree: a group that is entirely correct or entirely incorr

ReMem: Rethinking Perception and Memory in Long-Context Recommendation Agents

arXiv:2609.37311v1 Announce Type: new Abstract: Recent Recommendation Agents (RecAgents) offer a promising alternative by shifting recommendation to an active, user-side paradigm, where generative age

agentsarxiv-cs-ai

Applications

Rethinking Multimodal Fake News Detection in the Generative AI Era

arXiv:2609.36850v1 Announce Type: new Abstract: Generative content is increasingly entering the production and dissemination of news, transforming fake news from manually fabricated or simply manipula

applicationsarxiv-cs-cl

Same Bytes, Different Authority: Reserved-Token Representations in Chat-Template Prompt Injection

arXiv:2609.35932v1 Announce Type: cross Abstract: Prompt injection against LLM agents becomes much stronger when the injected instruction is wrapped in the model's own chat template. A forged template

Semantic Map Sharing and Capability-Aware Coverage Planning for AI-Native 6G Robotic Coordination

arXiv:2609.37666v1 Announce Type: cross Abstract: Search and Rescue (SAR) operations increasingly deploy heterogeneous teams of aerial and ground robots. However, conventional coverage methods typical

SIPO: Unifying Reinforcement Learning with On-Policy Self-Distillation

arXiv:2609.36742v1 Announce Type: new Abstract: Reinforcement learning with verifiable rewards (RLVR) has become a standard paradigm for improving large language models (LLMs) on various tasks, yet it

SkillWeaver: Agentic Exploration over Neural Interaction Skills for Scalable Robot Data Generation

arXiv:2609.36171v1 Announce Type: new Abstract: Large-scale demonstrations have driven unprecedented progress in robot learning, yet collecting robot data through teleoperation is expensive and diffic

agentsarxiv-cs-ro

ThinkingGuard: Decoding Implicit Hazards via Step-by-Step Risk Attribution in Multimodal Large Language Models

arXiv:2609.36562v1 Announce Type: cross Abstract: While Multimodal Large Language Models (MLLMs) are increasingly deployed in safety-critical domains, their reliability is threatened by multimodal imp

Timelines just got shorter... AI will understand the physical 3D space better than humans by November*

model-releasesemad-mostaque--x

Timelines just got shorter... AI will understand the physical 3D space better than humans by November* Is Google so back? Their new Gemini 4 Argon understands the physical 3D world better than any oth

Where Does Staleness Accumulate? Pool Aware Effective Staleness Control for Asynchronous RL in LLM Post-Training

arXiv:2609.36830v1 Announce Type: new Abstract: Fully asynchronous reinforcement learning (RL) improves resource utilization in large language model post-training by overlapping rollout generation wit

50B+ MoEs with few active parameters, what's the sweet spot for intelligence, agent speed, and affordable fine-tuning?

model-releasesr-localllama

I’m building a Polish General purpose legal Model that drafts documents, answers questions using legal sources, and has enough coding ability to handle some automation. The workflow is very tool-heavy

29 Sept 2026

AutoRef: Harness Optimization for Agentic Multi-Reference Image Generation

model-releasesarxiv-cs-cv

arXiv:2609.35530v1 Announce Type: new Abstract: Recent image generation models can take multiple reference images as input and combine them into a new image. However, multi-reference image generation

29 Sept 2026

Beyond Token Scale: Chunk-Level Sparse Autoencoders for Reliable Semantic Feature Discovery