每日论文 Daily Papers

来自 Hugging Face 与研究社区 · 中英对照

Daily Weekly Monthly

‹ 2026-08-17 ~ 2026-08-23 ›

跳转日期： 09-11 30 09-09 10 09-08 26 09-07 31 09-04 31 09-03 26 08-30 2 08-29 21 08-28 32 08-27 23 08-26 32 08-25 21 08-23 11 08-21 33 08-19 59 08-18 31 08-15 30 08-14 23 08-11 30 08-07 47 08-06 36 08-05 37 08-04 25 08-02 38 07-31 17 07-30 23 07-29 27 07-28 16 07-25 21 07-24 19

Can We Defend Against AI-Generated Video Attacks on Real-World Crisis Events? A Systematic Evaluation of Detectors, Generators and Social Dissemination

▲ 257 💬 1 ★ 36

SemComp-Bench: Benchmarking Semantic Task Completion in Video Generation

▲ 151 💬 4 ★ 11

EnvHarness: Awakening Static Worlds for Agent Learning

▲ 150 💬 1 ★ 8

Self-Supervised Visual On-Policy Distillation

▲ 142 💬 3

Zetta ζ: An Efficient Closed-Loop Embodied Harness for Self-Evolving Physical Intelligence

▲ 129 💬 1 ★ 186

SemaPLC: A Project-Grounded, Verification-Gated Agent Harness for PLC Code Generation

▲ 110 💬 1 ★ 32

HarnessEval-W: Agentifying the Evaluation of Visual Worlds

▲ 108 💬 1 ★ 132

Co-RL: Unsupervised Reasoning Emerges from Diverse Cohort in Multi-agent RL

▲ 80 💬 1 ★ 2

VibeWorlding: Can Multimodal Agents Construct 3D Open Worlds End-to-End?

▲ 50 💬 1 ★ 48

Large Discovery Models: Empirically-grounded Model-Based Open-Ended Search

▲ 47 💬 2 ★ 5

FACET: Preserving Source Intent and Executable State in Terminal Task Synthesis

▲ 47 💬 2 ★ 8

Learn What's Left, Not What's Mastered: Saturation Aware Advantage Reweighting for Multi-Reward Policy Optimization

▲ 43 💬 2

SPADE: Self-Play in Adaptive Synthetic Executable Environments

▲ 40 💬 1 ★ 3

Beyond Final Scores: A Systematic Evaluation of Agents for Long-Horizon AI Research and Development

▲ 39 💬 1

MOSS-VL Technical Report

▲ 37 💬 1 ★ 430

ASI-Bench: At the Dawn of Artificial Superintelligence

▲ 37 💬 2 ★ 10

4DAnyone: Create Anyone in 4D from a Casual Monocular Video

▲ 37 💬 3 ★ 20

FreeToken: Efficient Edge-Native MoE Serving with Bandwidth-Adaptive Execution

▲ 34 💬 1 ★ 18

ClawGym II: Exploring Black-Box RL on Agent Harness

▲ 33 💬 1 ★ 35

UI-Mate: Advancing Open-Weight Foundation GUI Agents with In-Context Demonstrations

▲ 32 💬 1 ★ 39

Intern-S2-Mobius: Foundation Model with Decoupled Knowledge and Reasoning

▲ 30 💬 1

WithEveryone: Unified Planning and Identity Grounding for Group Image Generation

▲ 30 💬 1 ★ 4

Training Chemical Plausibility-Aware Large Language Models for Single-Step Retrosynthesis

▲ 29 💬 1

SWE-bench Science: Can Coding Agents Resolve Engineering Tasks in Science?

▲ 29 💬 1 ★ 13

StateM: Reaching 95.3% Raw Accuracy, or a \$15 Frontier Run, on Terminal-Bench 2.1 via Harness Scaling

▲ 28 💬 1 ★ 49

Agentic ESOpt: Fine-Tuning Long-Horizon LLM Agents with Minimal GPU Requirements

▲ 24 💬 1 ★ 3

An Empirical Study of Training Pixel-Space Text-to-Image Diffusion Models

▲ 23 💬 1

MemTrapBench: Benchmarking Cognitive Traps in LLM Memory Use

▲ 23 💬 1 ★ 1

Marionette: Predicting World States, Rendering Geometry, Painting Appearance

▲ 22 💬 1

Advancing Open and Reproducible Relational Learning: RelArena-α, TabPFN-Rel and RPI

▲ 21 💬 1 ★ 18

Agentic Transaction: Towards ACID-Compliant Agent Systems

▲ 21 💬 1 ★ 6

How Do Agents Fail on AutoResearch: End-to-End Diagnostic Evaluation on 100 Real-World Frontier Research Tasks

▲ 21 💬 2

SimpleOPD: Simple Tokenizer-Agnostic On-Policy Distillation for Long-Context Reasoning

▲ 20 💬 1 ★ 3

MobileMem: Learning from a Year of Mobile Experiences

▲ 20 💬 1 ★ 12

Apodex Discovery: Reality Benchmarks and Environments for Evaluating and Building Discoverative Artificial Intelligence

▲ 20 💬 1

Embodied-Navigator: Point, Think, Memorize, and Align for Efficient Navigation

▲ 20 💬 1 ★ 208

DFM Mimir v1: An Open HRM Delivering Frontier Performance at 1B Parameters Using Only Permissible Post-Training Data

▲ 19 💬 1

GenRouter: Unified Workflow Routing for Agentic Image Generation

▲ 19 💬 1 ★ 22

Training Leaves Traces: Centered Residual Signatures for Language Model Lineage Verification

▲ 16 💬 1

HumanTracker: Towards Comprehensive and Human-Aligned Motion Tracking Benchmark

▲ 15 💬 1 ★ 26

Understanding Cognition-Induced Risks in Agentic AI Systems

▲ 15 💬 1

MegaParts: Scaling Part-Aware 3D Object Generation to 300 Parts via Token-Efficient Autoregressive Modeling

▲ 15 💬 1 ★ 513

Decision-Metric Alignment in Latent World Models: Diagnostics and Action-Conditioned Objectives for MPC Planning

▲ 15 💬 1

Looped Language Models Improve Compositional Tool Calling

▲ 15

GRNEdit: Efficient General Video Editing from a New Binary-Evidence Perspective in Generative Refinement Networks

▲ 13 💬 1 ★ 4

FM-Bench: A Benchmark for Long-Horizon Management with Competing Agents

▲ 13 ★ 13

Scaling Creative Writing Beyond Story-Centric Data with Attribute-Guided Genre Expansion

▲ 13 💬 1

Thinking in a Low-Resource Language: What SFT Builds, What RL Fixes, What Accuracy Cannot See

▲ 13 💬 3

CPI-Bench: A Comprehensive,Practical and Intelligent Benchmark for Real-World Image Editing

▲ 12 💬 1 ★ 6

Ventor-QTest: Threat-Model-Driven Verification of Vendor-Hosted LLM APIs

▲ 12 💬 1

SoftVTBench: A Deformation-Aware Visuo-Tactile Dataset and Benchmark for Deformable-Object Manipulation

▲ 12 💬 1

The More Popular, The Harder to Forget: Adaptive Popularity for LLM Unlearning

▲ 12 💬 1

R^3-Bench: LLMs Struggle with Resource-Rational Reasoning under Shared Budgets

▲ 11 💬 1 ★ 2

Agent Lightning v1.0: Towards Harnessed Agentic RL

▲ 11 💬 1

ForgeWM: Progressive Causal Training for Few-Step Action-Conditioned Video World Models

▲ 11 💬 1 ★ 76

PRM-as-a-Judge 1.5: A Toolkit for Robot Process Assessment

▲ 10 💬 1 ★ 5

Improving the matrix multiplication exponent with modern optimization and AlphaEvolve

▲ 10

Latent On-Policy Self-Distillation

▲ 9 💬 1 ★ 31

Claim-Level Reliability Assessment for Efficient Test-Time Reasoning

▲ 9 💬 1 ★ 2

Modular Cognitive Architecture Emerges in Large Language Models

▲ 9 💬 1 ★ 69

TRACE-Bench: Decomposing and Diagnosing Multi-Reference Image Generation

▲ 9 💬 1

VideoGAIA: A Benchmark for General AI Assistants on Agentic Video Understanding

▲ 9 💬 1 ★ 5

CoinVE-200K: A Large-Scale High-Quality Dataset for Compositional Instruction-Guided Video Editing

▲ 9 💬 1 ★ 3

Harness the Memory: A Holistic Evaluation of Memory Substrates in Memory Agents

▲ 9 💬 1

Hierarchical Self-Improvement: A Framework for Task-Specific Evolvable Agent Harnesses

▲ 9 💬 2 ★ 7

PACE-Bench: Benchmarking Physics Adaptation via Code Evolution in Dynamic Environments

▲ 8 💬 1 ★ 2

From Corpora to Co-Evolving Capabilities: Capability-Centric Data Design for Generalist Image Generation

▲ 8 💬 1

TinyCast: Probabilistic Zero-Shot Forecasting with Computed Periodicity

▲ 8 💬 2 ★ 2

NARU: A Benchmark for NARrative Evolution and Cultural Nuance Understanding in Japanese Extreme Long Video

▲ 8 💬 3 ★ 3

Forecast Collapse in Time-Series Foundation Models

▲ 7 💬 1

Second Thought: Reasoning in Parallel as LLM Agents Act and Observe

▲ 7 💬 2

Multimodal Model Diffing for Feature Discovery and Control

▲ 7 💬 1

τ_0-VLA: a Hierarchical Robot Foundation Model with World-Model-Guided Test-Time Computation

▲ 7 💬 2 ★ 518

HarnessRisk: A Lifecycle-Oriented Benchmark for Agent Harness Safety

▲ 6 💬 1 ★ 5

From Sequence to Structure: Relational Uncertainty Propagation for LLM Agents

▲ 6 💬 1 ★ 1

SkillEvo: Self-Renewing Evolution Gradients from Multi-Turn Interaction Feedback

▲ 6 💬 2

GOAG: Generative and Object-Agnostic Grasp Planner for Dexterous Robotic Manipulation

▲ 6 💬 2 ★ 3

CoToGrasp: Contact-Topology-Conditioned Dexterous Grasp Synthesis via Canonical Workspace Learning

▲ 6 💬 2 ★ 4

The Embedder's Dilemma: LLMs Are Better, but at What Cost?

▲ 6 💬 2

QuoteBench: How Matched Scores Can Hide Command-Path Failures

▲ 6 💬 2 ★ 3

Scaling Domain Data Repetition in LLM Pretraining

▲ 5 💬 1

LittleLearner: Language Models Under Pedagogically Controlled Knowledge Exposure

▲ 5 💬 1

AnyTalk: Speech Animation for Arbitrary Characters Leveraging a Video Generation Model

▲ 5 💬 1 ★ 1

NaviDC-OCR: Navigating Document Parsing Across Digital and Camera-Captured Documents

▲ 5 💬 1 ★ 20

Repo0: Design-Driven Zero-to-All Code Generation

▲ 5 💬 1

Listening Forward: Next Patch Embedding Prediction Enables Scalable Audio Learners

▲ 5 💬 1 ★ 2

Agents Catching Agents: Shortcut Cascades and Benchmark Gaming in Clinical Multi-Agent Systems

▲ 4 💬 1 ★ 2

Verifier-Induced Support Reshaping in On-Policy Optimization

▲ 4 💬 1 ★ 2

ConceptFormer: Learning Adaptive Latent Concepts for Query-Document Alignment in Visual Document Retrieval

▲ 4 💬 1 ★ 3

ENTLORE: A Graph-Grounded Benchmark for Latent Organizational Reasoning in Enterprise Question Answering

▲ 4 💬 1 ★ 2

Abra: Scaling Diffusion Image Training

▲ 4

Energy-Guided Flow Matching

▲ 4 💬 1 ★ 15

FlashPrefill V2: Block-Sparse Prefill Attention for Long-Context LLM Serving

▲ 4 💬 1 ★ 1

Dion3: Full-Stack Orthogonal Updates

▲ 3 💬 1

UniProbe: A Learnable Token-Level Hallucination Detector for Large VLMs using Multi-Structural Internal Representations

▲ 3 💬 1

Prior Audit-Repair Context Shifts LLM Verifier Thresholds Toward Leniency

▲ 3 💬 1

GS-Voxel: Fitting-Free Structured Latents for Large-Scale 3DGS Generation

▲ 3 💬 2

MathForm: Scaling Mathematical Autoformalization with Knowledge Retrieval and Verification-Guided Refinement

▲ 3 💬 1 ★ 3

Inject, Align, Recover: Staged Post-Training for Retrieval-Free Document Knowledge Internalization

▲ 3 💬 1

EXIMO: VLM Guided Exploration of VLA Policies

▲ 3

A Pathway to General-Purpose Scientific AI: Multimodal Comprehension of Scientific Images

▲ 2 💬 1

Nanbeige4.2-3B on Apple Silicon: Fixing Deployment Bugs and Decreasing Looped Transformer Memory Overhead

▲ 2 💬 1 ★ 1

Amplified Does Not Mean Predictive: Reasoning Behaviors in Thinking Models

▲ 2 💬 1

Drive, Pack, Fly: The Travelling Thief Problem with Drone

▲ 2 💬 1 ★ 3

Gathered, Not Admitted: How Attention Brings a Latent Variable into Verbalizable Form

▲ 2 💬 1

DumpsterCluster: From Dumpster Diving to Serving LLaMA-70B on $60 GPUs

▲ 2 💬 1

Plausible but Not Valid: A Psychometric Audit of LLMs as Synthetic Survey Respondents

▲ 2 💬 1

Security Assessment of DeepSeek Harness with A.I.G: Evaluating Resistance to Indirect Prompt Injection

▲ 2 💬 1

StartupBench: Benchmarking General-Purpose Agents on Market-Validated End-to-End Workflows

▲ 2

aDSL: Agentic 3D Creation via Joint Agent-Program Design

▲ 2

Dynamic Multi-Byte Prediction With Hierarchical Language Models

▲ 2 💬 1

Personalized Auto-Research: Towards a True AI Co-Scientist

▲ 2 💬 1

SkillGate: Training In-Policy Skill Selection in Long-Horizon Agents

▲ 2 💬 2

LLMs Get Smarter from Targeted Synthetic Multilingual Data

▲ 2 💬 1

Bounded Agents: Delegation Security for Multi-Agent AI Systems

▲ 2 💬 1 ★ 2

Temporal Multi-Signal Fusion for Token-Level Hallucination Detection

▲ 2 💬 1

Towards Quantifying Benchmark Optimization in ASR Models

▲ 2 💬 1 ★ 1

PolicyGuide: From Guarding One Action to Guiding the Whole Workflow for Policy-Compliant LLM Agents

▲ 2 💬 1

Chain-of-Experience for Continual LLM Improvement

▲ 2 💬 1

FlowEvo: Self-Evolving Agents through the Co-Evolution of Workflows and Executable Skills

▲ 2 💬 2 ★ 7

SPARGen: Unifying Spatial Perception and Reasoning through Native Multimodal Generation

▲ 1 💬 1

Who Speaks Matters: Authority-Aware Multi-View RAG over Italian Parliamentary Proceedings

▲ 1 💬 1 ★ 2

UNMASK: Discovering and Causally Verifying Spurious Shortcuts in Text Classifiers

▲ 1 💬 2

A Plug-and-Play 2D Motion Interface for Real-World Motion Language Models

▲ 1 💬 1 ★ 1

WorldRover: A Scalable Synthetic Video Data Engine for World Exploration with Rich Annotations

▲ 1 💬 1

When Context Bites: Detecting RAG Poisoning via Document-Level Attention Collapse

▲ 1 💬 1

Position: AI Agents in Scientific Teams Should Be Studied as Human-Agent Systems

▲ 1 💬 1

Valid Per-Field Selective Risk Control for Document Extraction: Three Failure Modes, a Validity Ladder, and When Conditioning Pays

▲ 1 💬 1

Unifying Graph Neural Networks Through a Common Layer Equation

▲ 1 💬 1

Generation as Auxiliary Supervision: Enhancing Visual Understanding at Zero Inference Overhead via Decoupled Embedding Prediction

▲ 0

StreamOPD: A Post-Training Recipe with Spatio-Temporal Cue Gating for Streaming Video Understanding

▲ 0 💬 1 ★ 4

HiFi-BRep: High-Fidelity Latent Representation for Robust B-Rep Generation

▲ 0 💬 1 ★ 3

Prototype-Rectified Iterative Self-supervised Manifold Denoising under Severe Acoustic Shift

▲ 0 💬 1 ★ 1

SPK: Eliciting Structured Prior Knowledge for Interpretable Out-of-Distribution Detection in Real-Time Object Detection

▲ 0 💬 1