Tencent Hunyuan · 66 records

dateidtitleauthorsvenue↑
2026-11-172611.20882Accelerate Bench: Evaluating Agents for Application-Level Program OptimizationBowen Liu et al.arXiv18
2026-11-052611.06518EmoReward: Self-Evolving Reward Calibration for Emotional Video Generation RLMingzhe Zheng et al.arXiv25
2026-11-052611.08240FewStepRL: Fine-tuning Few-step Diffusion Models with Reinforcement LearningHaitao Lin et al.arXiv9
2026-10-272610.36729Hy-Nomas: Towards Free-form Interactive World GenerationQiaoling Chen et al.arXiv122
2026-10-212610.27435Harness Self-Evolution for Long-Horizon Terminal AgentsRuhan Wang et al.arXiv13
2026-10-192610.23181Tree-Aware Posterior Distillation for Long-Horizon Agentic Reinforcement LearningHeming Zou et al.arXiv53
2026-10-142610.18092LifeSearch: Benchmarking Deep Research Agents for Real-World Life and ProductivityYibo Wang et al.arXiv17
2026-10-142610.04906Scaling Verifiable Environments for Long-horizon Work AgentsJiazheng Zhang et al.arXiv5
2026-10-142609.37377Beyond Prompt Count: How Data Shapes Transfer in On-Policy DistillationJiaxuan Wang et al.arXiv3
2026-10-092610.06801MC-Sparse: Deconstructing and Closing the Dense-Sparse Attention Gap in Diffusion TransformersJiarui Chen et al.arXiv40
2026-10-092610.12458OmniCapBench: A Deep-Structured Evaluation Framework for Fine-Grained Audio-Visual CaptioningZhongyu Yang et al.arXiv15
2026-10-092609.34344Learning to Steer, Steering to See: Unveiling the Geometry of RLVR in Large Language Models via Trainable VectorsYuchen Cai et al.arXiv4
2026-10-062610.05416Prism: Dynamic Sparse Attention for Native 2K Joint Video-Audio Generation Model TrainingShuyuan Tu et al.arXiv17
2026-10-052610.02826Scaling Trajectories for Complex Tasks through Recursive Self-RewriteZongxia Li et al.arXiv42
2026-10-022610.01215AutoGUIWorld: Image Generators as Visual World Models for GUI AgentCheng Yang et al.arXiv38
2026-10-022609.34621Does Native 3D Texture Generation Necessarily Require 3D Assets for Training?Jiangshan Wang et al.arXiv3
2026-09-292609.35457How Far Are We from Removing the Visual Encoder? Scaling Laws for Encoder-Free Multimodal PretrainingLin Chen et al.arXiv61
2026-09-252609.30199ExplorationBench: Measuring AI Systems' Exploration in Verifiable Alien WorldsMing Zhang et al.arXiv30
2026-09-242609.27284Hunyuan-A13B Technical ReportTencent Hunyuan Team et al.arXiv21
2026-09-142609.13141SAS: Simple Attention Sparsification via End-to-End Optimization of Context RankingZhiwei Li et al.arXiv68
2026-09-102609.11042T1: Terminal Agent Reinforcement Learning for Long-Horizon TasksJunyao Yang et al.arXiv65
2026-09-092609.08936AuK Technical Report: An Open-Source Foundational Model for Speech Generation and EditingZiyang Ma et al.arXiv138
2026-09-092609.08977Omni Interaction Agent Technical ReportOrantqing et al.arXiv73
2026-09-082609.03241FlowBalance: Verifier-Grounded Self-Improvement from On-Policy Reasoning ExperienceZixun Huang et al.arXiv43
2026-09-042609.04128Environment Evolution for Terminal AgentsZhiyuan Fan et al.arXiv10
2026-09-022609.00646DramaChain Bench: An End-to-End Benchmark for Short-Drama GenerationHaoyuan Shi et al.arXiv15
2026-09-012608.29252Dynamic Important Example Mining for Reinforcement FinetuningHaoru Tan et al.arXiv5
2026-08-262608.24794CAFE: Self-Improving Search Agents Need Co-Evolving FeedbackBoyang Liu et al.arXiv6
2026-08-252608.21833GameXpert-Bench: How Far Are Coding Agents from Expert Game Development?Kun Chen et al.arXiv18
2026-08-212608.20336WithEveryone: Unified Planning and Identity Grounding for Group Image GenerationHengyuan Xu et al.arXiv39
2026-08-182608.15930UI-Mate: Advancing Open-Weight Foundation GUI Agents with In-Context DemonstrationsZihan Ding et al.arXiv48
2026-08-072608.05248WorldClaw: Agentic 3D Open-World Generation at ScaleChunchao Guo et al.arXiv66
2026-08-062608.05466Recursive Synthesis for Long-Horizon Terminal TasksZhongzhi Li et al.arXiv80
2026-08-052608.02711Hunyuan3D-Buffalo 1.0: A Unified Multimodal Model for Scalable 3D Generation, Understanding, and EditingJunliang Ye et al.arXiv46
2026-07-272607.22043Scaling Native Multimodal Pre-Training From ScratchHaoyuan Wu et al.arXiv28
2026-07-222607.18722Stale but Stable: Staleness-Adaptive Trust Regions for Stabilizing Asynchronous Reinforcement LearningJunyao Yang et al.arXiv33
2026-07-172607.15273MeanFlowNFT: Bringing Forward-Process RL to Average-Velocity GeneratorsYushi Huang et al.arXiv12
2026-07-162607.13285Harness Handbook: Making Evolving Agent Harnesses Readable,Navigable, and EditableRuhan Wang et al.arXiv115
2026-07-132607.08964Long-Horizon-Terminal-Bench: Testing the Limits of Agents on Long-Horizon Terminal Tasks with Dense Reward-Based GradingZongxia Li et al.arXiv72
2026-07-082607.02980Hierarchical Sparse Attention Done Right: Toward Infinite Context ModelingXiang Hu et al.arXiv82
2026-07-082607.05804TurnOPD: Making On-Policy Distillation Turn-Aware for Efficient Long-Horizon Agent TrainingYuhang Zhou et al.arXiv20
2026-07-032607.02291Optimizing Visual Generative Models via Distribution-wise RewardsRuihang Li et al.arXiv16
2026-07-032606.27669When Search Agents Should Ask: DiscoBench for Clarification-Aware Deep SearchYiling Tao et al.arXiv11
2026-07-012606.32039GEAR: Guided End-to-End AutoRegression for Image SynthesisBin Lin et al.arXiv15
2026-07-012606.30673PolyFlow: Continuous Topology Embedding Flow Matching for Artist-style Mesh GenerationChunshi Wang et al.arXiv11
2026-06-302606.29705GUICrafter: Weakly-Supervised GUI Agent Leveraging Massive Unannotated ScreenshotsSunqi Fan et al.arXiv14
2026-06-262606.27313ViQ: Text-Aligned Visual Quantized Representations at Any ResolutionXumin Yu et al.arXiv28
2026-06-242606.23543VeriEvol: Scaling Multimodal Mathematical Reasoning via Verifiable Evol-InstructHaoling Li et al.arXiv3
2026-06-232606.23049Training Open Models for Agentic Phone UseZhengyang Tang et al.arXiv18
2026-06-232606.14971FastMix: Fast Data Mixture Optimization via Gradient DescentHaoru Tan et al.arXiv4
2026-06-182606.19236STARE: Surprisal-Guided Token-Level Advantage Reweighting for Policy Entropy StabilityHaipeng Luo et al.arXiv5
2026-06-152606.14409Hy-Embodied-0.5-VLA: From Vision-Language-Action Models to a Real-World Robot Learning StackHe Zhang et al.arXiv15
2026-06-082606.06891Stream3D-VLM: Online 3D Spatial Understanding with Incremental Geometry PriorsHanxun Yu et al.arXiv5
2026-05-292605.30248GenClaw: Code-Driven Agentic Image GenerationJunyan Ye et al.arXiv38
2026-05-282605.28548GEM: Generative Supervision Helps Embodied IntelligenceRuowen Zhao et al.arXiv28
2026-05-262605.26108Reinforcing Few-step Generators via Reward-Tilted Distribution MatchingYushi Huang et al.arXiv5
2026-05-212605.20873PlanningBench: Generating Scalable and Verifiable Planning Data for Evaluating and Training Large Language ModelsZiliang Zhao et al.arXiv42
2026-05-182605.11739Learning to Foresee: Unveiling the Unlocking Efficiency of On-Policy DistillationYuchen Cai et al.arXiv15
2026-05-182605.15876Unlocking Dense Metric Depth Estimation in VLMsHanxun Yu et al.arXiv11
2026-05-132605.11711Debiased Model-based Representations for Sample-efficient Continuous ControlJiafei Lyu et al.arXiv7
2026-05-132605.04702FaithfulFaces: Pose-Faithful Facial Identity Preservation for Text-to-Video GenerationYuanzhi Wang et al.arXiv1
2026-05-132605.07545Implicit Preference Alignment for Human Image AnimationYuanzhi Wang et al.arXiv1
2026-05-122605.09262Reinforcing Multimodal Reasoning Against Visual DegradationRui Liu et al.arXiv5
2026-05-122605.09269DeltaRubric: Generative Multimodal Reward Modeling via Joint Planning and VerificationRui Liu et al.arXiv4
2026-05-112605.06139Listwise Policy Optimization: Group-based RLVR as Target-Projection on the LLM Response SimplexYun Qu et al.arXiv73
2026-05-072605.05185OpenSearch-VL: An Open Recipe for Frontier Multimodal Search AgentsShuang Chen et al.arXiv53