repoGitHubTrust 82 · PrimaryPublished 1mo agoLive · 22d ago

bytedance/deer-flow

An open-source long-horizon SuperAgent harness that researches, codes, and creates. With the help of sandboxes, memories, tools, skill, subagents and message gateway, it handles different levels of tasks that could take minutes to hours.

Lineage graph

Paper → model → repo connections mined from source citations (Tier-1 exact match).

Why these links exist

Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.

FuzzyOverlapping authors or contributors · 62%Wan-Dancer: A Hierarchical Framework for Minute-scale Coherent Music-to-Dance Generation →
“Shared author/contributor keys: wang”
FuzzyOverlapping authors or contributors · 62%Experience Memory Graph: One-Shot Error Correction for Agents →
“Shared author/contributor keys: wang”
FuzzyOverlapping authors or contributors · 62%Do Agent Optimizers Compound? A Continual-Learning Evaluation on Terminal-Bench 2.0 →
“Shared author/contributor keys: wang”
FuzzyOverlapping authors or contributors · 62%Simon-SR: Spatially Adaptive Modulation and Visual Prompt Adaptation for Text-Reinforced Super-Resolution →
“Shared author/contributor keys: wang”
FuzzyOverlapping authors or contributors · 62%Self-Evolving Agent Harnesses via Gated Semantic Quality-Diversity →
“Shared author/contributor keys: wang”
FuzzyOverlapping authors or contributors · 62%From RGB Generation to Dense Field Readout: Pixel-Space Dense Prediction with Text-to-Image Models →
“Shared author/contributor keys: wang”
FuzzyOverlapping authors or contributors · 62%DevicesWorld: Benchmarking Cross-Device Agents in Heterogeneous Environments →
“Shared author/contributor keys: wang”
FuzzyOverlapping authors or contributors · 62%TRACE: Turn-level Reward Assignment via Credit Estimation for Long-Horizon Agents →
“Shared author/contributor keys: wang”

Implements

paperWan-Dancer: A Hierarchical Framework for Minute-scale Coherent Music-to-Dance Generation paperExperience Memory Graph: One-Shot Error Correction for Agents paperDo Agent Optimizers Compound? A Continual-Learning Evaluation on Terminal-Bench 2.0 paperSimon-SR: Spatially Adaptive Modulation and Visual Prompt Adaptation for Text-Reinforced Super-Resolution paperSelf-Evolving Agent Harnesses via Gated Semantic Quality-Diversity paperFrom RGB Generation to Dense Field Readout: Pixel-Space Dense Prediction with Text-to-Image Models paperDevicesWorld: Benchmarking Cross-Device Agents in Heterogeneous Environments paperTRACE: Turn-level Reward Assignment via Credit Estimation for Long-Horizon Agents paperM$^\text{4}$World: A Multi-view Multimodal Driving World Model for Interactive Object Manipulation and Minute-long Streaming paperAspectCLIP: Optimizing CLIP Representation Space via Aspect-Guided Consistency Regularization paperTerraZero: Procedural Driving Simulation for Zero-Demonstration Self-Play at Scale paperLearning Mechanistic Reasoning for Chemical Reactions with Large Language Models paperScalable Visual Pretraining for Language Intelligence paperLLM Agents for Deliberative Collaboration: A Study on Joint Decision Making Under Partial Observability paperBeyond Isolated Objects: Relationship-aware Open Vocabulary Scene Understanding via 3D Scene Graph Analysis paperAn Efficient vLLM-Based Inference Pipeline for Unified Audio Understanding and Generation paperDKCD: Domain Knowledge-Enhanced Causal Discovery from Unstructured Data paperCan Induced Emotion Bias LLM Behaviors in Sequential Decision Making?paperMachine Learning for Depression Screening and Intervention: an Original Circadian Rhythm Score-based Methodology paperBiSCo-LLM: Lookup-Free Binary Spherical Coding for Extreme Low-Bit Large Language Model Compression paperText-Driven 3D Indoor Scene Synthesis in Non-Manhattan Environments paperCheckRLM: Effective Knowledge-Thought Coherence Checking in Retrieval-Augmented Reasoning paperSPEARBench: A Benchmark for Naturalness Evaluation in Streaming Speech-to-Speech Language Models paper4DR360: State Reasoning for Joint 3D Detection and Occupancy Prediction in 4D Radar-Camera Full-Scene Perception paperSecure Decentralized Federated Learning via Gossip and Virtual Voting paperWhen Local Monitors Miss Compositional Harm: Diagnosing Distributed Backdoors in Multi-Agent Systems paperToken-Flow Firewall: Semantic Runtime Auditing for Persistent AI Agents paperCortex: A Bidirectionally Aligned Embodied Agent Framework for Long-horizon Manipulation paperMedRealMM: A Real-World Multimodal Benchmark for Chinese Online Medical Consultation paperUniClawBench: A Universal Benchmark for Proactive Agents on Real-World Tasks paperDisciplineGen-1M: A Large-Scale Dataset for Multidisciplinary Visual Generation and Editing paperPhysMani: Physics-principled 3D World Model for Dynamic Object Manipulation paperSynthetic-to-Real Translation for Class-Agnostic Motion Prediction paperCoRe: A Comprehensive Framework for Cross-Image Comparative Reasoning in Vision-Language Models paperInside the Unfair Judge: A Mechanistic Interpretability Account of LLM-as-Judge Bias paperAir Quality Downscaling with Station-Guided Pseudo-Supervision paperAVSCap: Orchestrating Audio-Visual Synergy for Omni-modal Video Captioning paperGaP: A Graph-as-Policy Multi-Agent Self-Learning Harness For Variational Automation Tasks paperAccelerated Mixing Time of Randomized Hamiltonian Monte Carlo paperFinding H. pylori in the Fine Print: Evidence-Linked Multi-Agent Case Finding from Gastric Biopsy Reports paperPurified OPSD: On-Policy Self-Distillation Without Losing How to Think paperInformation Gain-based Rollout Policy Optimization: An Adaptive Tree-Structured Rollout Approach for Multi-Turn LLM Agents paperOptimizing Visual Generative Models via Distribution-wise Rewards paperCurateEvo: Data-Curation Evolving for Agentic Post-Training paperReCal3R: Reliability-Calibrated Learning Rates for Streaming 3D Reconstruction paperFrom SRA to Self-Flow: Data Augmentation or Self-Supervision?paperIdeas Have Genomes: Benchmarking Scientific Lineage Reasoning and Lineage-Grounded Idea Generation paperDT-Guard: Intent-Driven Reasoning-Active Training for Reasoning-Free LLM Safety Guardrail paperTowards Mechanistically Understanding Why Memorized Knowledge Fails to Generalize in Large Language Model Finetuning paperDynamic Neural Graph Encoding of Inference Processes in Deep Weight Space paperUniVR: Thinking in Visual Space for Unified Visual Reasoning paperControllable Sim Agents with Behavior Latents paperWhen Does Tool Use Increase the Expressive Power of Finite-Precision Recurrent Models?paperHy-Embodied-VLM-1.0: Efficient Physical-World Agents paperCausalMix: Data Mixture as Causal Inference for Language Model Training paperFADRA: Frequency-Aware Diffusion with Residual Adaptation for Video Face Restoration paperUltraX: Refining Pre-Training Data at Scale with Adaptive Programmatic Editing paperWhat Images Cannot Say: Language-Guided Olfactory Representation Learning paperLean-QIT: Towards a Formal Infrastructure for Quantum Information Theory paperDemonstrating TOFFEE: A Learned System for Synthesizing Data Agent Trajectories at Scale paperDo Vision-Language-Action Models Mean What They Say? On the Role of Faithfulness in Embodied Reasoning paperUnderstanding the Robustness of Distributed Self-Supervised Learning Frameworks Against Non-IID Data paperVoxENES 2026: Benchmarking Generalization of Speech Spoofing Detectors Against LLM-Era TTS and Voice Conversion paperPixWorld: Unifying 3D Scene Generation and Reconstruction in Pixel Space paperZO-Act: Efficient Zeroth-Order Fine-Tuning via One-Shot Activation-Informed Low-Rank Subspaces paperMIRAGE: Defending Long-Form RAG Against Misinformation Pollution paperScaling Behavior Foundation Model for Humanoid Robots paperBrainPilot: Automating Brain Discovery with Agentic Research paperLQCDMaster: Agentic Scientific Computing for Lattice Quantum Chromodynamics Research paperRubrics on Trial: Evolving Rubrics from a Single Query via Synthetic Pairwise Evidence paperBadWAM: When World-Action Models Dream Right but Act Wrong paperBenchmarking Multimodal Large Language Models for Scientific Visualization Literacy paperDAPGNet: Dynamic Adaptive Physics-Guided Graph Diffusion Network for Hyperspectral Image Classification paperRoGS: Adaptive Meshgrid Gaussian for Large-Scale Road Surface Mapping paperVideo = World + Event Stream paperJADE-GS: Joint Alternating Deblurring Guided by Events in 3D Gaussian Splatting paperFrom Draft to Draft-Free: One-Step Video Object Removal via Privileged Distillation and Fast Planting paperVideoChat3: Fully Open Video MLLM for Efficient and Generalist Video Understanding paperSciDiagramEdit: Learning to Edit Scientific Diagrams from Paper Revisions paperMotion-Conditioned Multi-View Fusion for Myocardial Infarction Localization from Echocardiography paperStructural-Semantic Reciprocal Learning for Unsupervised Visible-Infrared Person Re-Identification paperWhen Words Are Safe But Actions Kill: Probing Physical Danger Beyond Text Safety in Hidden-State Risk Space paperDetecting Answer-Driven Reasoning in LLM-Based Educational Tutors via Truncated Chain-of-Thought Auditing paperTowards Precision Therapy in Hepatocellular Carcinoma: A Clinical-Reasoning LLM for Risk Stratification and Treatment Guidance paperAlayaWorld: Long-Horizon and Playable Video World Generation paperWho Grades the Grader? Co-Evolving Evaluation Metrics and Skills for Self-Improving LLM Agents paperWorldDirector: Building Controllable World Simulators with Persistent Dynamic Memory paperWebSwarm: Recursive Multi-Agent Orchestration for Deep-and-Wide Web Search paperCycle-World: Mitigating Error Accumulation in Long-term Video World Models via Reverse-Prediction Cycle Consistency paperAn Exam for Active Observers paperContextual Semantic Relevance Tracks fMRI BOLD Responses During Naturalistic Speech Comprehension paperBetter Starts, Better Ends: Bootstrapped Iterative Self-Reasoning Distillation for Compressed Reasoning paperAn MLIR-Based Compilation Method for Large Language Models paperPerceive-to-Reason: Decoupling Perception and Reasoning for Fine-Grained Visual Reasoning paperJoyNexus: Service-Oriented Multi-Tenant Post-Training for VLA Models paperCRAFT: Clustering Rubrics to Diagnose Weak LLM Capabilities and Generate Targeted Fine-Tuning Data paperSearching Videos as Trees: Self-Correcting Agents for Grounded Long Video QA paperArtChart: A Benchmark for Faithful Artistic Chart Generation with Integrated Text Rendering paperDPNeXt: A Lightweight Multi-Scale Feature Fusion Framework for Efficient ViT-Based Multi-Task Dense Prediction paperBeyond Unfolding: 60x Faster One-Stage Unmixing for Closely-Spaced Infrared Small Targets paperStraight-Path Flow Matching for Incomplete Multi-View Clustering paperGeoMix: Descriptor-Free Visual Localization via Global Context and Multi-Detector Training paperGeometry and Gradient-based Partitioning for Panoramic Outdoor Reconstruction paperPaperRouter-Agent: A Content-Grounded LLM Agent for Personalized Hierarchical Paper Routing paperActive rejection enables reliable generalization of universal machine-learning interatomic potentials paperLenGuard-GPC: Length Guarding with Guided-Prompt Consistency for Spatial Reasoning Reinforce Learning paperDistilled Reinforcement Learning for LLM Post-training paperNoise-Robust Box-Supervised Infrared Small Target Detection via Physics-Inspired Soft Label Optimization paperDepthART: Scaling Foundation Monocular Depth to Tiny Models paperToward Anthropomorphic Dialogue: A Closed-Loop Framework for Human-Like Chat Generation, Evaluation, and Preference Alignment paperHow Jailbreak Attacks Inform Safety Alignment: A Defender-Centric, Shapley-Based Evaluation of Jailbreak Contributions paperSynH-Rank: Quality-Aware Code Search via Diverse Data Synthesis and Hierarchical Ranking Training paperCross-Branch Conflict as a Shield: Safeguarding Facial Identities in Unified Multimodal Image Editing paperSTBridge: Shared-Target Alignment for Bridging Understanding and Generation in UMMs paperAdvSerial: Physical Adversarial Attacks on Infrastructure-mounted Pedestrian Detectors via Semantic Feature Suppression paperALLUDE: A Unified Evaluation System for Configurable Attacks in Differentiable Environments paperAutoregressive B-Rep Shape Generation with Parametric Surfaces paperSparse Evidence Can Suffice: Agentic Evidence Seeking for Multimodal Video Misinformation Detection paperSWE-Pruner Pro: The Coder LLM Already Knows What to Prune paperSelectInfer: Selective Neuron Loading and Computation for On-Device LLMs paperGigaPath-Flash and GigaTIME-Flash: Efficient Pathology Foundation Models for Whole-Slide and Tumor Microenvironment Analysis paperEnhancing Rubric-based RL via Self-Distillation paperEVOLVE: Efficient Learned Volume Compression with Variable-Rate Encoding on a Cross-Domain Database paperAn Early Warning of Emerging Biosecurity Risks in Frontier LLMs paperThe Many Senses of Visual Similarity: A Text-Prompted Image Perceptual Metric paperLossless-INR: Lossless Volumetric Implicit Neural Representations paperO-VAD: Industrial Video Anomaly Detection through Object-Centric Tracking and Reasoning paperOcclusion-Aware Panoptic Segmentation with Joint Position Embedding and Occlusion-Level Attention paperVGOcc: Learning Visual-Geometric Gaussians for Vision-Centric 3D Driving Occupancy Prediction paperUniETP: Unifying Environments for Generalizable Embodied Task Planning paperWorldCupArena: Fine-Grained Evaluation of Language Models and Deep-Research Agents on Football Forecasting paperAutoMem: Automated Learning of Memory as a Cognitive Skill paperMBTI: A Multi-Branch Efficient Fine-Tuning Framework for Hyperspectral Image Classification with Foundation Models paperSMetric: Rethink LLM Scheduling for Serving Agents with Balanced Session-centric Scheduling paperVaseMuseum: Digital Intelligent Museum for Ancient Greek Pottery paperBioSecBench-Surveillance: A Verifiable Benchmark for AI Agents in Pathogen Genomic Surveillance paperCodeRescue: Budget-Calibrated Recovery Routing for Coding Agents paperSequential Learner Modeling Using Multi-Relational Graph Convolutional Networks paperCopy Less, Ground More: Overcoming Repetitive Copying in Long-Context Reasoning via Evidence-Aware Reinforcement Learning paperMeetingToM: Evaluating Multimodal LLMs on Theory-of-Mind Reasoning in Multi-Party Meetings paperDAIS: Dependency-Aware Intermediate QA Supervision for Complex Reasoning paperMedDDC-Eval: Diagnosis-Decoupled Evaluation of Multi-Turn Medical Consultation Agents paperISO: An RLVR-Native Optimization Stack paperOmniReasoner: Thinking with Long Audio-Video via Native Tool Use paperExpertVerse: A General-Purpose Benchmark for Expert-Level Reasoning in Knowledge-Intensive Visual Synthesis paperNo Training, Better Flights: Test-Time Scaled VLMs for UAV Navigation paperCR-Refiner: An Object-Centric Optimal Transport Reranker for Edit-Conditioned 3D Scene Retrieval paperGATE-3D: Geometry-Aware Test-time Adaptive Reranking for Open-Set 3D Shape Retrieval paperBeyond the Single Camera: Agentic Multi-View Reasoning in Sports Video Understanding paperEdgeBench: Unveiling Scaling Laws of Learning from Real-World Environments paperThe Geometry of Memorization: Finite-Time Spectral Sensitivity as a Diagnostic for Flow Matching Models paperCAAD: Causality-Aware Multivariate Time Series Anomaly Detection via Multi-Scale Alignment and Structural Causal Consistency paperXALPHA: A Memory-Driven AI Quant Researcher for Hypothesis-to-Code Alpha Discovery paperDSpark: Confidence-Scheduled Speculative Decoding with Semi-Autoregressive Generation paperAudio-Zero: Label-Free Self-Evolution for Fine-Grained Audio Reasoning paperOn the Systematic Challenges of Culturally Loaded Machine Translation: Dream of the Red Chamber as the Cultural Lens paperLabel-Free Finite-Volume-Residual Training of Attention Graph Neural Networks for Coupled Thermo-Fluid Fields paperPhaseAware: Interpretable Human-in-the-Loop Rehabilitation Scoring with Boundary Monitoring paperPercepCap: Video Captioner with Structured Spatio-Temporal Perception paperOLEDLM: A Unified Language Model for OLED Molecular Design paperLook Less, Think Faster: Joint Token-Compute Adaptation for Multimodal LLMs paperEvolving Cache Schedules for Fast Diffusion Policy Inference paperVera: Identity-Faithful Human Subject-to-Video Generation paperRS-RIE-Bench: Benchmarking Reasoning-Guided Remote Sensing Image Editing paperPerceptDrive: Perception Prior World-Action Modeling with Adaptive Expert Routing for End-to-End Autonomous Driving paperRIM: A Retrieval-In-Matching Framework for Cross-Domain Global Visual Localization of UAVs paperGaussianSeed: Hierarchical Gaussian Seeding for High-Resolution 3D Occupancy Prediction paperPushing the Frontier of Full-Song Generation: Hierarchical Autoregressive Planning Meets Flow-Matching Rendering paperExtraGS: Enhancing Endoscopic View Extrapolation via Diffusion-Guided 3D Gaussian Splatting paperAlphaOracle: Oracle bone script decipherment via human-workflow-inspired deep learning paperGeoSearcher: Anchor-Guided Progressive Reasoning for Remote Sensing Visual Grounding with Process Supervision paperBefore the Action: Benchmarking LLMs on Prospective Hypothesis Discovery paperAREX: Towards a Recursively Self-Improving Agent for Deep Research paperPATS: Policy-Aware Training Scaffolding for Agentic Reinforcement Learning paperAdaptive Depth Sparse Framework: Similarity-Driven Resource Allocation for Pre-Trained LLMs paperOne More Turn, Less Regret: A Regret-Based Multi-Turn Benchmark for LLMs' Clarification Policies paperToward Generalizable Cognitive Impairment Detection with Speech-Based Multimodal Large Language Models paperTexture++: Elevating 3D Asset Texture Resolution with a Region-Aware Diffusion Model paperTowards Privacy-Preserving Federated Prompt Tuning under Data Heterogeneity: A Subspace-Decomposed Expert Approach paperASTRA-Net: Anatomy-Specific Transfer and Representation Alignment for Drug-Induced Sleep Endoscopy Segmentation paperFlash EQ-Linear: Accelerating Equivariant Linear Layers via Group-wise Discrete Fourier Transform paperThe Dark Room in the Reward Channel: Dense Prediction Rewards Collapse GRPO-Trained LLM Agents -- and What Actually Works paperBeyond Sycophancy: Structured Resistance and Compliance in LLM Moral Reasoning paperZero-Flow Two-Sample Tests

Covers (incoming)

newsByteDance Enters Physical AI Arena: Unlocking the Second Wave of the Tech Industry Feast - 36 Kr

Related across the graph

paperSimon-SR: Spatially Adaptive Modulation and Visual Prompt Adaptation for Text-Reinforced Super-Resolution personJasonOA888 paperWan-Dancer: A Hierarchical Framework for Minute-scale Coherent Music-to-Dance Generation paperFrom RGB Generation to Dense Field Readout: Pixel-Space Dense Prediction with Text-to-Image Models paperDKCD: Domain Knowledge-Enhanced Causal Discovery from Unstructured Data paperTerraZero: Procedural Driving Simulation for Zero-Demonstration Self-Play at Scale paperScalable Visual Pretraining for Language Intelligence personforeleven paperText-Driven 3D Indoor Scene Synthesis in Non-Manhattan Environments paperLabel-Free Finite-Volume-Residual Training of Attention Graph Neural Networks for Coupled Thermo-Fluid Fields paperBiSCo-LLM: Lookup-Free Binary Spherical Coding for Extreme Low-Bit Large Language Model Compression paperAlphaOracle: Oracle bone script decipherment via human-workflow-inspired deep learning paperSelectInfer: Selective Neuron Loading and Computation for On-Device LLMs paperCheckRLM: Effective Knowledge-Thought Coherence Checking in Retrieval-Augmented Reasoning paperOmniReasoner: Thinking with Long Audio-Video via Native Tool Use paperHow Jailbreak Attacks Inform Safety Alignment: A Defender-Centric, Shapley-Based Evaluation of Jailbreak Contributions paperSPEARBench: A Benchmark for Naturalness Evaluation in Streaming Speech-to-Speech Language Models paper4DR360: State Reasoning for Joint 3D Detection and Occupancy Prediction in 4D Radar-Camera Full-Scene Perception paperSecure Decentralized Federated Learning via Gossip and Virtual Voting paperThe Many Senses of Visual Similarity: A Text-Prompted Image Perceptual Metric paperWhen Local Monitors Miss Compositional Harm: Diagnosing Distributed Backdoors in Multi-Agent Systems paperMachine Learning for Depression Screening and Intervention: an Original Circadian Rhythm Score-based Methodology paperNoise-Robust Box-Supervised Infrared Small Target Detection via Physics-Inspired Soft Label Optimization paperCortex: A Bidirectionally Aligned Embodied Agent Framework for Long-horizon Manipulation paperGaussianSeed: Hierarchical Gaussian Seeding for High-Resolution 3D Occupancy Prediction paperAspectCLIP: Optimizing CLIP Representation Space via Aspect-Guided Consistency Regularization paperBefore the Action: Benchmarking LLMs on Prospective Hypothesis Discovery paperMedRealMM: A Real-World Multimodal Benchmark for Chinese Online Medical Consultation paperCan Induced Emotion Bias LLM Behaviors in Sequential Decision Making?paperDisciplineGen-1M: A Large-Scale Dataset for Multidisciplinary Visual Generation and Editing personknuknY paperPhysMani: Physics-principled 3D World Model for Dynamic Object Manipulation paperISO: An RLVR-Native Optimization Stack paperSelf-Evolving Agent Harnesses via Gated Semantic Quality-Diversity paperTowards Privacy-Preserving Federated Prompt Tuning under Data Heterogeneity: A Subspace-Decomposed Expert Approach paperTexture++: Elevating 3D Asset Texture Resolution with a Region-Aware Diffusion Model paperEvolving Cache Schedules for Fast Diffusion Policy Inference paperCross-Branch Conflict as a Shield: Safeguarding Facial Identities in Unified Multimodal Image Editing paperAn Exam for Active Observers paperBetter Starts, Better Ends: Bootstrapped Iterative Self-Reasoning Distillation for Compressed Reasoning paperInside the Unfair Judge: A Mechanistic Interpretability Account of LLM-as-Judge Bias paperAir Quality Downscaling with Station-Guided Pseudo-Supervision paperMotion-Conditioned Multi-View Fusion for Myocardial Infarction Localization from Echocardiography paperUniClawBench: A Universal Benchmark for Proactive Agents on Real-World Tasks personsyzhang622 paperLearning Mechanistic Reasoning for Chemical Reactions with Large Language Models paperCoRe: A Comprehensive Framework for Cross-Image Comparative Reasoning in Vision-Language Models paperGaP: A Graph-as-Policy Multi-Agent Self-Learning Harness For Variational Automation Tasks paperBrainPilot: Automating Brain Discovery with Agentic Research paperContextual Semantic Relevance Tracks fMRI BOLD Responses During Naturalistic Speech Comprehension paperAccelerated Mixing Time of Randomized Hamiltonian Monte Carlo paperPerceptDrive: Perception Prior World-Action Modeling with Adaptive Expert Routing for End-to-End Autonomous Driving paperAn Efficient vLLM-Based Inference Pipeline for Unified Audio Understanding and Generation paperStructural-Semantic Reciprocal Learning for Unsupervised Visible-Infrared Person Re-Identification paperFinding H. pylori in the Fine Print: Evidence-Linked Multi-Agent Case Finding from Gastric Biopsy Reports paperSynthetic-to-Real Translation for Class-Agnostic Motion Prediction paperLLM Agents for Deliberative Collaboration: A Study on Joint Decision Making Under Partial Observability paperSMetric: Rethink LLM Scheduling for Serving Agents with Balanced Session-centric Scheduling persongreatmengqi paperInformation Gain-based Rollout Policy Optimization: An Adaptive Tree-Structured Rollout Approach for Multi-Turn LLM Agents paperOLEDLM: A Unified Language Model for OLED Molecular Design paperDPNeXt: A Lightweight Multi-Scale Feature Fusion Framework for Efficient ViT-Based Multi-Task Dense Prediction paperBeyond Isolated Objects: Relationship-aware Open Vocabulary Scene Understanding via 3D Scene Graph Analysis paperToken-Flow Firewall: Semantic Runtime Auditing for Persistent AI Agents paperVera: Identity-Faithful Human Subject-to-Video Generation paperLook Less, Think Faster: Joint Token-Compute Adaptation for Multimodal LLMs paperJoyNexus: Service-Oriented Multi-Tenant Post-Training for VLA Models personyangzheli paperOcclusion-Aware Panoptic Segmentation with Joint Position Embedding and Occlusion-Level Attention paperOptimizing Visual Generative Models via Distribution-wise Rewards paperCurateEvo: Data-Curation Evolving for Agentic Post-Training paperReCal3R: Reliability-Calibrated Learning Rates for Streaming 3D Reconstruction paperFrom SRA to Self-Flow: Data Augmentation or Self-Supervision?paperIdeas Have Genomes: Benchmarking Scientific Lineage Reasoning and Lineage-Grounded Idea Generation paperDT-Guard: Intent-Driven Reasoning-Active Training for Reasoning-Free LLM Safety Guardrail paperO-VAD: Industrial Video Anomaly Detection through Object-Centric Tracking and Reasoning paperBadWAM: When World-Action Models Dream Right but Act Wrong paperAudio-Zero: Label-Free Self-Evolution for Fine-Grained Audio Reasoning paperFlash EQ-Linear: Accelerating Equivariant Linear Layers via Group-wise Discrete Fourier Transform paperUniETP: Unifying Environments for Generalizable Embodied Task Planning paperTowards Mechanistically Understanding Why Memorized Knowledge Fails to Generalize in Large Language Model Finetuning paperDynamic Neural Graph Encoding of Inference Processes in Deep Weight Space paperAVSCap: Orchestrating Audio-Visual Synergy for Omni-modal Video Captioning paperWorldCupArena: Fine-Grained Evaluation of Language Models and Deep-Research Agents on Football Forecasting personhetaoBackend paperBenchmarking Multimodal Large Language Models for Scientific Visualization Literacy paperHy-Embodied-VLM-1.0: Efficient Physical-World Agents paperEVOLVE: Efficient Learned Volume Compression with Variable-Rate Encoding on a Cross-Domain Database paperDistilled Reinforcement Learning for LLM Post-training paperPercepCap: Video Captioner with Structured Spatio-Temporal Perception paperPurified OPSD: On-Policy Self-Distillation Without Losing How to Think paperUniVR: Thinking in Visual Space for Unified Visual Reasoning paperGeometry and Gradient-based Partitioning for Panoramic Outdoor Reconstruction paperPaperRouter-Agent: A Content-Grounded LLM Agent for Personalized Hierarchical Paper Routing personggnnggez paperAdvSerial: Physical Adversarial Attacks on Infrastructure-mounted Pedestrian Detectors via Semantic Feature Suppression personhenry-byted paperAlayaWorld: Long-Horizon and Playable Video World Generation paperCausalMix: Data Mixture as Causal Inference for Language Model Training paperOn the Systematic Challenges of Culturally Loaded Machine Translation: Dream of the Red Chamber as the Cultural Lens paperBeyond the Single Camera: Agentic Multi-View Reasoning in Sports Video Understanding paperControllable Sim Agents with Behavior Latents paperWhen Does Tool Use Increase the Expressive Power of Finite-Precision Recurrent Models?paperPhaseAware: Interpretable Human-in-the-Loop Rehabilitation Scoring with Boundary Monitoring paperWho Grades the Grader? Co-Evolving Evaluation Metrics and Skills for Self-Improving LLM Agents paperFADRA: Frequency-Aware Diffusion with Residual Adaptation for Video Face Restoration paperZO-Act: Efficient Zeroth-Order Fine-Tuning via One-Shot Activation-Informed Low-Rank Subspaces paperWebSwarm: Recursive Multi-Agent Orchestration for Deep-and-Wide Web Search paperSequential Learner Modeling Using Multi-Relational Graph Convolutional Networks paperDo Vision-Language-Action Models Mean What They Say? On the Role of Faithfulness in Embodied Reasoning paperNo Training, Better Flights: Test-Time Scaled VLMs for UAV Navigation paperUnderstanding the Robustness of Distributed Self-Supervised Learning Frameworks Against Non-IID Data personHuixin615 paperMBTI: A Multi-Branch Efficient Fine-Tuning Framework for Hyperspectral Image Classification with Foundation Models paperWhen Words Are Safe But Actions Kill: Probing Physical Danger Beyond Text Safety in Hidden-State Risk Space paperJADE-GS: Joint Alternating Deblurring Guided by Events in 3D Gaussian Splatting paperRoGS: Adaptive Meshgrid Gaussian for Large-Scale Road Surface Mapping paperGigaPath-Flash and GigaTIME-Flash: Efficient Pathology Foundation Models for Whole-Slide and Tumor Microenvironment Analysis paperBioSecBench-Surveillance: A Verifiable Benchmark for AI Agents in Pathogen Genomic Surveillance paperWorldDirector: Building Controllable World Simulators with Persistent Dynamic Memory paperPATS: Policy-Aware Training Scaffolding for Agentic Reinforcement Learning paperDetecting Answer-Driven Reasoning in LLM-Based Educational Tutors via Truncated Chain-of-Thought Auditing paperUltraX: Refining Pre-Training Data at Scale with Adaptive Programmatic Editing person18062706139fcz paperLQCDMaster: Agentic Scientific Computing for Lattice Quantum Chromodynamics Research paperRS-RIE-Bench: Benchmarking Reasoning-Guided Remote Sensing Image Editing paperAn Early Warning of Emerging Biosecurity Risks in Frontier LLMs persondependabot[bot]paperOne More Turn, Less Regret: A Regret-Based Multi-Turn Benchmark for LLMs' Clarification Policies personShenAC-SAC paperWhat Images Cannot Say: Language-Guided Olfactory Representation Learning paperLean-QIT: Towards a Formal Infrastructure for Quantum Information Theory paperVideo = World + Event Stream paperTowards Precision Therapy in Hepatocellular Carcinoma: A Clinical-Reasoning LLM for Risk Stratification and Treatment Guidance paperDemonstrating TOFFEE: A Learned System for Synthesizing Data Agent Trajectories at Scale paperDAIS: Dependency-Aware Intermediate QA Supervision for Complex Reasoning paperFrom Draft to Draft-Free: One-Step Video Object Removal via Privileged Distillation and Fast Planting paperCodeRescue: Budget-Calibrated Recovery Routing for Coding Agents paperVGOcc: Learning Visual-Geometric Gaussians for Vision-Centric 3D Driving Occupancy Prediction paperScaling Behavior Foundation Model for Humanoid Robots paperDSpark: Confidence-Scheduled Speculative Decoding with Semi-Autoregressive Generation paperXALPHA: A Memory-Driven AI Quant Researcher for Hypothesis-to-Code Alpha Discovery paperM$^\text{4}$World: A Multi-view Multimodal Driving World Model for Interactive Object Manipulation and Minute-long Streaming paperThe Geometry of Memorization: Finite-Time Spectral Sensitivity as a Diagnostic for Flow Matching Models paperAutoMem: Automated Learning of Memory as a Cognitive Skill paperEnhancing Rubric-based RL via Self-Distillation paperEdgeBench: Unveiling Scaling Laws of Learning from Real-World Environments paperPixWorld: Unifying 3D Scene Generation and Reconstruction in Pixel Space paperGATE-3D: Geometry-Aware Test-time Adaptive Reranking for Open-Set 3D Shape Retrieval paperCAAD: Causality-Aware Multivariate Time Series Anomaly Detection via Multi-Scale Alignment and Structural Causal Consistency paperGeoSearcher: Anchor-Guided Progressive Reasoning for Remote Sensing Visual Grounding with Process Supervision paperExpertVerse: A General-Purpose Benchmark for Expert-Level Reasoning in Knowledge-Intensive Visual Synthesis paperBeyond Unfolding: 60x Faster One-Stage Unmixing for Closely-Spaced Infrared Small Targets paperStraight-Path Flow Matching for Incomplete Multi-View Clustering paperGeoMix: Descriptor-Free Visual Localization via Global Context and Multi-Detector Training paperExtraGS: Enhancing Endoscopic View Extrapolation via Diffusion-Guided 3D Gaussian Splatting paperVaseMuseum: Digital Intelligent Museum for Ancient Greek Pottery paperActive rejection enables reliable generalization of universal machine-learning interatomic potentials paperAREX: Towards a Recursively Self-Improving Agent for Deep Research paperLenGuard-GPC: Length Guarding with Guided-Prompt Consistency for Spatial Reasoning Reinforce Learning paperArtChart: A Benchmark for Faithful Artistic Chart Generation with Integrated Text Rendering paperAn MLIR-Based Compilation Method for Large Language Models paperCRAFT: Clustering Rubrics to Diagnose Weak LLM Capabilities and Generate Targeted Fine-Tuning Data paperPerceive-to-Reason: Decoupling Perception and Reasoning for Fine-Grained Visual Reasoning personEilen6316 paperLossless-INR: Lossless Volumetric Implicit Neural Representations paperCycle-World: Mitigating Error Accumulation in Long-term Video World Models via Reverse-Prediction Cycle Consistency paperMedDDC-Eval: Diagnosis-Decoupled Evaluation of Multi-Turn Medical Consultation Agents paperZero-Flow Two-Sample Tests paperVideoChat3: Fully Open Video MLLM for Efficient and Generalist Video Understanding paperBeyond Sycophancy: Structured Resistance and Compliance in LLM Moral Reasoning paperExperience Memory Graph: One-Shot Error Correction for Agents paperSTBridge: Shared-Target Alignment for Bridging Understanding and Generation in UMMs personMagicCube paperVoxENES 2026: Benchmarking Generalization of Speech Spoofing Detectors Against LLM-Era TTS and Voice Conversion personforx11 paperALLUDE: A Unified Evaluation System for Configurable Attacks in Differentiable Environments paperToward Generalizable Cognitive Impairment Detection with Speech-Based Multimodal Large Language Models paperSparse Evidence Can Suffice: Agentic Evidence Seeking for Multimodal Video Misinformation Detection personocto-patch paperASTRA-Net: Anatomy-Specific Transfer and Representation Alignment for Drug-Induced Sleep Endoscopy Segmentation paperCopy Less, Ground More: Overcoming Repetitive Copying in Long-Context Reasoning via Evidence-Aware Reinforcement Learning paperSynH-Rank: Quality-Aware Code Search via Diverse Data Synthesis and Hierarchical Ranking Training personWillemJiang personGujiassh paperSearching Videos as Trees: Self-Correcting Agents for Grounded Long Video QA personly-wang19 paperMeetingToM: Evaluating Multimodal LLMs on Theory-of-Mind Reasoning in Multi-Party Meetings personwhhe newsByteDance Enters Physical AI Arena: Unlocking the Second Wave of the Tech Industry Feast - 36 Kr paperRubrics on Trial: Evolving Rubrics from a Single Query via Synthetic Pairwise Evidence paperToward Anthropomorphic Dialogue: A Closed-Loop Framework for Human-Like Chat Generation, Evaluation, and Preference Alignment paperCR-Refiner: An Object-Centric Optimal Transport Reranker for Edit-Conditioned 3D Scene Retrieval paperAutoregressive B-Rep Shape Generation with Parametric Surfaces paperDepthART: Scaling Foundation Monocular Depth to Tiny Models paperSWE-Pruner Pro: The Coder LLM Already Knows What to Prune personfancyboi999 paperPushing the Frontier of Full-Song Generation: Hierarchical Autoregressive Planning Meets Flow-Matching Rendering paperRIM: A Retrieval-In-Matching Framework for Cross-Domain Global Visual Localization of UAVs paperThe Dark Room in the Reward Channel: Dense Prediction Rewards Collapse GRPO-Trained LLM Agents -- and What Actually Works paperDAPGNet: Dynamic Adaptive Physics-Guided Graph Diffusion Network for Hyperspectral Image Classification paperTRACE: Turn-level Reward Assignment via Credit Estimation for Long-Horizon Agents personLofiSu paperAdaptive Depth Sparse Framework: Similarity-Driven Resource Allocation for Pre-Trained LLMs personxunliu paperDo Agent Optimizers Compound? A Continual-Learning Evaluation on Terminal-Bench 2.0 paperSciDiagramEdit: Learning to Edit Scientific Diagrams from Paper Revisions personLittleChenLiya paperDevicesWorld: Benchmarking Cross-Device Agents in Heterogeneous Environments paperMIRAGE: Defending Long-Form RAG Against Misinformation Pollution

Topics