Tsinghua University · 47 records

dateidtitleauthorsvenue↑
2026-11-162611.21366ALIVE: Making Streaming Video Models InteractiveJinzhao Li et al.arXiv40
2026-11-102611.14033Few-Step Streaming Video Generation via Condition ShiftingDeyuan Liu et al.arXiv88
2026-11-062611.06500Tools Need Not Speak in Text: Latent Context Compression for AgentsYuetong Liu et al.arXiv3
2026-10-30W7224806679Long-term electricity load forecasting model based on multi-period and multi-scale analysisXiaolin Zhang et al.PLoS ONE
2026-10-292610.39497Does Scaling Toward AGI Depend on What System-1 Lets System-2 Execute?Qi Wang et al.arXiv61
2026-10-292610.40599HairDeck: Animation-Ready Stylized Hair Cards from a Single ImageYuze He et al.arXiv23
2026-10-272610.33413AV-FABLE: Audio-Visual Manipulation Forensics in Long-Form VideosYuesheng Huang et al.arXiv43
2026-10-212610.22444Event-STVG: Spatio-Temporal Video Grounding for Long-Form Event ReasoningYizhe Zhang et al.arXiv17
2026-10-212610.30132LongZero: Enhancing Long-Context Reasoning in LLMs from Zero DataZhennan Wan et al.arXiv8
2026-10-162610.22355OmniVerifier-C1: Decomposed Graded Verifier for Rule-Governed Generative World ModelingBowei Liu et al.arXiv35
2026-10-152610.20828CoTranslator: Reverse to Forward Chain-of-Thought Translator for Reasoning DistillationZinan Tang et al.arXiv7
2026-10-12W7147307924Thermal conductivity and temperature-induced bandgap renormalization in crystalline and amorphous Ga2O3Rustam I. Arabov et al.Applied Physics Letters
2026-10-012609.39685RoboCoach: World Models as Active Coaches for Compositional Robot SkillsJiajun Liu et al.arXiv11
2026-09-302609.35514One Proposal for Every Margin: Zero-Shot Amortized Sequential Importance Sampling for Binary MatricesRuishuo Chen et al.arXiv3
2026-09-192609.22966Transferring the Intelligence of VLMs to Robotic ControlMeng-Hao Guo et al.arXiv67
2026-09-172609.18779CERA-MoA: Co-Evolving Routing Mechanisms with Continually Learning LLM AgentsJiaxuan Jiang et al.arXiv15
2026-09-162609.15982The Router Within: Eliciting Native Skill Routing from a Frozen LLMRuishuo Chen et al.arXiv7
2026-09-152609.11638Vidu S2: Real-Time Interactive, Editable, and Spatial Video GenerationJintao Zhang et al.arXiv237
2026-09-032608.21450Beyond Visual Similarity: Entity-Aligned Retrieval for Knowledge-Based Visual Question AnsweringHangrui Xu et al.arXiv10
2026-08-182608.13900Agentic Transaction: Towards ACID-Compliant Agent SystemsZhaoyan Sun et al.arXiv28
2026-08-102608.04569Relevant but Incomplete: Referential Dangling as a Paradigm-Level Failure Mode in Hard Prompt CompressionZhengpei Hu et al.arXiv12
2026-08-102608.05597Uncertainty-Aware World Model for Aerial Image-Goal NavigationDeyi Zhu et al.arXiv10
2026-08-072608.05987AgentOPSD: Recursive Self-Distillation for Agentic Reinforcement LearningZi-Han Wang et al.arXiv82
2026-08-072608.06146PaDoc: Layout-Grounded Parallel Decoding for Document ParsingHao Yu et al.arXiv12
2026-08-052608.02791Better, Stronger, Faster, and Broader: Structured All-Mask Prediction for MLLM-Based SegmentationJiazhen Liu et al.arXiv6
2026-08-052608.00730Push-Wiper: Toward General-Purpose Robotic Cleaning across Varied Stains and Surfaces with Segmented Pushing TrajectoriesRenhao Lu et al.arXiv5
2026-08-052607.28993ST-WAM: Semantic-Temporal World Action Model for Robust Manipulation under Visual Distribution ShiftsMingxin Wang et al.arXiv3
2026-07-102607.03118Vidu S1: A Real-Time Interactive Video Generation ModelJintao Zhang et al.arXiv146
2026-07-032607.01647AgenticDataBench: A Comprehensive Benchmark for Data AgentsZhaoyan Sun et al.arXiv35
2026-07-022606.19271TurboServe: Serving Streaming Video Generation Efficiently and EconomicallyYouhe Jiang et al.arXiv34
2026-07-022607.01104CausalMix: Data Mixture as Causal Inference for Language Model TrainingZinan Tang et al.arXiv19
2026-06-302606.26740LiveEdit: Towards Real-Time Diffusion-Based Streaming Video EditingXinyu Wang et al.arXiv82
2026-06-092606.09669SpatialWorld: Benchmarking Interactive Spatial Reasoning of Multimodal Agents in Real-World TasksHongcheng Gao et al.arXiv50
2026-06-082603.16142Parametric Social Identity Injection and Diversification in Public Opinion SimulationHexi Wang et al.arXiv1
2026-06-042606.02684Filter, Then Reweight: Rethinking Optimization Granularity in On-Policy DistillationYuying Li et al.arXiv16
2026-06-042606.03895Agent libOS: A Library-OS-Inspired Runtime for Long-Running, Capability-Controlled LLM AgentsYingqi ZhangarXiv3
2026-06-032606.04923Reproducing, Analyzing, and Detecting Reward Hacking in Rubric-Based Reinforcement LearningXuekang Wang et al.arXiv42
2026-06-012605.23458One-Forcing: Towards Stable One-Step Autoregressive Video GenerationJiaqi Feng et al.arXiv4
2026-05-272605.25437Does Seeing More Mean Knowing More? Mono-Anchored Advantage Normalization for Multi-Source Visual ReasoningFanhu Zeng et al.arXiv10
2026-05-262605.25160SimuWoB: Simulating Real-World Mobile Apps for Fast and Faithful GUI Agent BenchmarkingGuohong Liu et al.arXiv6
2026-05-222605.22538Segment Anything with Motion, Geometry, and Semantic Adaptation for Complex Nonlinear Visual Object TrackingDeyi Zhu et al.arXiv4
2026-05-212602.07892Safety Alignment as Continual Learning: Mitigating the Alignment Tax via Orthogonal Gradient ProjectionGuanglong Sun et al.arXiv2
2026-05-192605.14278KVPO: ODE-Native GRPO for Autoregressive Video Alignment via KV Semantic ExplorationRuicheng Zhang et al.arXiv36
2026-05-192605.18643Post-Trained MoE Can Skip Half Experts via Self-DistillationXingtai Lv et al.arXiv30
2026-05-142605.08978Learning to Explore: Scaling Agentic Reasoning via Exploration-Aware Policy OptimizationXingyuan Hua et al.arXiv1
2026-05-112605.059974DThinker: Thinking with 4D Imagery for Dynamic Spatial UnderstandingZhangquan Chen et al.arXiv16
2026-05-082605.04956KernelBench-X: A Comprehensive Benchmark for Evaluating LLM-Generated GPU KernelsHan Wang et al.arXiv7