ByteDance · 36 records

dateidtitleauthorsvenue↑
2026-11-052611.05609On the Generalization Limitation of On-Policy Distillation for Post-RL ConsolidationYichen Wang et al.arXiv23
2026-11-032611.04720OneTrans-GSR: A Generative Slate Ranking Paradigm Unifying Reranking and Mix-Ranking with One Autoregressive TransformerFeng Cheng et al.arXiv36
2026-10-232610.32253RSIHub: A Composable Framework for Agent-Evolution MethodsZimu Wang et al.arXiv49
2026-10-202610.26081DeOverlay: A Benchmark and Source-Anchored Flow for Video Overlay RemovalDeshui Miao et al.arXiv18
2026-10-202610.23935Factorization-Guided Training and Inference for Diffusion-Drafted Speculative DecodingJiakun Fan et al.arXiv11
2026-10-162610.19849DeltaStreamer: Towards Human-Like Agentic Streaming Video UnderstandingBoyuan Sun et al.arXiv38
2026-10-162610.21055Athanor: An Agent-Native System for Autonomous Post-Training ResearchYufa Zhou et al.arXiv16
2026-10-142609.38847Scoring Higher, Answering Worse: Mitigating Reward Hacking in Rubric-Based RL via Protocol-Level RubricsMaoqi Liu et al.arXiv18
2026-10-072610.03543DuoMatching: Joint-Marginal Distribution Matching for Few-Step Video GenerationJiahao Zhan et al.arXiv95
2026-10-072610.02188DMAD: Distribution Matching as Adversarial Distillation for Fast Visual GenerationZhengming Yu et al.arXiv4
2026-09-302609.38140Breaking the Uniformity Trap: Scaling Video Diffusion Model via SplitMoEYu Xu et al.arXiv5
2026-09-292609.33295TraceDance: An Automated System for Building Agent Behavior Benchmarks from Real-World Agent Deployment TracesDehai Min et al.arXiv70
2026-09-162609.15975Disentangling Representation Evolution in Transformers through Directional DecompositionShwai He et al.arXiv9
2026-09-102609.09187AgenticGen: Reward-Guided Agentic Video Generation for AdvertisingXingyuan Bu et al.arXiv13
2026-08-312608.18524DART-SD: Diamond-topology Aware Retrieval and Tuning for Self-Distillation of Multi-Turn Tool-Calling AgentsHangrui Xu et al.arXiv38
2026-08-172608.12209Generation as Auxiliary Supervision: Enhancing Visual Understanding at Zero Inference Overhead via Decoupled Embedding PredictionZhongbin Guo et al.arXiv5
2026-08-112608.09802SWE-Bench ProMax: Benchmarking Agents on Large-Scale Multilingual Code RefactoringYuling Shi et al.arXiv73
2026-08-102608.02148Douyin Multimodal Embedding Model Technical ReportHaonan Chen et al.arXiv14
2026-08-062608.03632When Teachers Mislead: Spurious-Signal-Aware On-Policy DistillationYinuo Jiang et al.arXiv19
2026-08-042608.02023SwanTale: Unified Multi-Speaker Speech and Audio Generation for Instruct and Zero-Shot TasksYu Zhang et al.arXiv66
2026-07-302607.25659CoRT: Counterfactual Replay for Token-Level Rubric-Guided Policy OptimizationBo-Wen Zhang et al.arXiv32
2026-07-242607.21017TableVerse: A Large-scale Tabletop Dataset with Real-world Grounded Layouts for Generalizable ManipulationBoyuan Wang et al.arXiv6
2026-07-212607.18213SWE-Pruner Pro: The Coder LLM Already Knows What to PruneYuhang Wang et al.arXiv46
2026-07-172607.12800UniVR: Thinking in Visual Space for Unified Visual ReasoningZhongwei Ren et al.arXiv29
2026-07-162607.13124ShortOPD: Recovering Pruned LLMs with Short-to-Long On-Policy DistillationQingyu Zhang et al.arXiv14
2026-07-012606.28436Dockerless: Environment-Free Program Verifier for Coding AgentsWenhao Zeng et al.arXiv78
2026-06-302606.29887SafePyramid: A Hierarchical Benchmark for In-context Policy GuardrailingJiacheng Zhang et al.arXiv3
2026-06-242606.24874FLUX3D: High-Fidelity 3D Gaussian Generation with Diffusion-Aligned Sparse RepresentationHaorui Ji et al.arXiv2
2026-06-222606.19534PerceptionDLM: Parallel Region Perception with Multimodal Diffusion Language ModelsYueyi Sun et al.arXiv63
2026-06-172606.17730ActWorld: From Explorable to Interactive World Model via Action-Aware MemoryZhexiao Xiong et al.arXiv9
2026-06-052606.06294Towards One-to-Many Temporal GroundingQi Xu et al.arXiv8
2026-06-012605.30993SwanVoice: Expressive Long-Form Zero-Shot Speech Synthesis for Both Monologue and DialogueRuiqi Li et al.arXiv54
2026-05-272605.27366MUSE-Autoskill: Self-Evolving Agents via Skill Creation, Memory, Management, and EvaluationHuawei Lin et al.arXiv28
2026-05-222605.22344Bernini: Latent Semantic Planning for Video DiffusionBernini Team et al.arXiv20
2026-05-152605.11458Adaptive Teacher Exposure for Self-Distillation in LLM ReasoningZihao Han et al.arXiv3
2026-05-042605.00809Let ViT Speak: Generative Language-Image Pre-trainingYan Fang et al.arXiv32