{"fields":["id","title","date","tags","disciplines","orgKey"],"count":311,"rows":[["2610.28632","Joints, Camera, Action! Co-Generating 3D Motion and Video for Arbitrary Rigs","2026-10-20",["visual-gen","vision"],["film-animation"],"nvidia"],["2610.28630","SolarRenderer: Few-Step Generative Reference Rendering","2026-10-20",["visual-gen","efficiency"],["software-engineering"],"show-lab"],["2610.28594","Learning the 4D World through Point Trajectory Forecasting","2026-10-20",["visual-gen","robotics"],["film-animation"],null],["2610.28578","Sage-A1.5: From Long-Horizon Execution to Scientific Exploration with Capability Composition","2026-10-20",["agent","ai4science"],["economics"],"intern-science"],["2610.28354","S-Agent 2.0: Spatial Reasoning with a General Coding Agent","2026-10-20",["agent","reasoning"],["software-engineering"],"ropedia"],["2610.27999","Matrix-Game 4.0: Harnessing Video Models for Real-Time Playable Game Worlds","2026-10-20",["visual-gen","agent"],["software-engineering"],null],["2610.27911","MMEnterprise-Bench: Evaluating Enterprise Agents over Evolving Multimodal Records","2026-10-20",["benchmark","agent"],["economics"],null],["2610.27628","PAR: Progress-Aligned Recurrence for Length Generalization in State Tracking","2026-10-20",["llm","reasoning"],["computer-science"],"ant-international"],["2610.26860","OpenVLHarness: Real World Visual Agentic Tool-Use with Multimodal Memory","2026-10-19",["multimodal","agent"],["computer-science"],"apple"],["2610.26855","Ropedia W0: Scaling Loco-Manipulation World Models across Embodiments with Human Experience","2026-10-19",["robotics","visual-gen"],["robotics"],"ropedia"],["2610.26853","InfiniDream: Open-Ended Photorealistic 3D Worlds Through Procedural Generation","2026-10-19",["vision","code"],["software-engineering"],"alaya-lab"],["2610.26851","Manga2Animate: An Agentic Framework for Faithful Manga Animation with Script and Storyboard Planning","2026-10-19",["agent","visual-gen"],["film-animation"],"national-yang-ming-chiao-tung-university"],["2610.26844","AnyGoal: On World Model Generalization Beyond Expert Rollouts","2026-10-19",["benchmark","agent"],["computer-science"],"galilai-group"],["2610.26837","SpecKCast: Spectral Kalman Forecasting for Diffusion Acceleration","2026-10-19",["visual-gen","efficiency"],["film-animation"],"samsung-research"],["2610.26834","Kronecker Delta Rule for Long Context Linear Attention","2026-10-19",["efficiency","llm"],["mathematics"],"nx-ai"],["2610.26825","MemT-1.0: Towards Memory-Native Evolution","2026-10-19",["llm","efficiency"],["computer-science"],"memtensor"],["2610.26824","SpecMRL: Redesign Matryoshka Representation Learning within Parametric Spectrum","2026-10-19",["efficiency"],["computer-science"],null],["2610.26802","Scaling Parameter-Efficient Fine-Tuning for Brain Encoding","2026-10-19",["ai4science","efficiency"],["neuroscience"],"mind-lab"],["2610.26770","HarnessPro: Recursive Harness Self-Improvement with Pareto Optimization for Performance-Cost Trade-offs","2026-10-19",["agent","code"],["computer-science"],null],["2610.26625","EgoXperience: Learning World-Grounded Egocentric Representations from Human Experience","2026-10-19",["vision","robotics","benchmark"],["computer-science"],"ropedia"],["2610.26549","Holi-World: Building Holistic World Intelligence from Human-Ego Experience","2026-10-19",["vision","benchmark"],["computer-science"],"ropedia"],["2610.26216","RoAR-VAE: Improving Latent Diffusability through Robust Prediction","2026-10-19",["visual-gen"],["computer-science"],"tsinghua-leaplab"],["2610.26203","Focused Omni-Modal GUI Agents","2026-10-19",["agent","multimodal"],["computer-science"],"nanyang-technological-university"],["2610.25950","AdaExploreBench: Benchmarking LLM Agents for Proactive Exploration and Test-time Adaptation in Evolving Environments","2026-10-19",["benchmark","agent"],["computer-science"],"hkust"],["2610.25844","SWIFT: Self-guided Weak-to-strong dIFfusion models via Target extrapolation","2026-10-19",["visual-gen"],["computer-science"],"alibaba"],["2610.25736","Learning from Fleet Experience: Collective Experience Learning Across 96 Robots","2026-10-19",["robotics"],["robotics"],null],["2610.25690","ScaleNav: Scalable Internet Video Pretraining for Language-Conditioned Local Navigation","2026-10-19",["robotics"],["robotics"],null],["2610.25404","MT Forcing: Towards Stable and Consistent Multi-Turn Image Editing with Autoregressive Diffusion","2026-10-18",["visual-gen"],["computer-science"],"sensenova"],["2610.25078","CoEvo-RSI: When Harnesses Evolve, Which Models Should Follow?","2026-10-18",["agent","llm"],["neuroscience"],null],["2610.25043","VLA-Rolling: Fewer Denoising Steps, Better Control via Adaptive Action Reuse","2026-10-18",["robotics","efficiency"],["robotics"],"zju-omniai"],["2610.24960","CoworkEnv: Synthesizing Training Environments for Long-Horizon Cowork Agents","2026-10-18",["agent","rl"],["computer-science"],null],["2610.24087","OR-License: Evaluating LLM Agents as Real-World Operations Research Practitioners","2026-10-16",["benchmark","agent"],["materials"],"data-science-artificial-intelligence-lab-at-hkust"],["2610.23787","DLite: History-free Speculative Decoding for Agentic AI","2026-10-16",["efficiency","agent"],["computer-science"],"shanghai-jiao-tong-university"],["2610.23751","MTEdit: Packed Trajectories Beat Rollouts for Multi-Turn Image Editing","2026-10-16",["visual-gen","benchmark"],["economics"],"sensenova"],["2610.23542","Execution Before Memory: Diagnosing Memory Failures in Video World Models","2026-10-16",["benchmark","visual-gen"],["computer-science"],"the-university-of-hong-kong"],["2610.23479","HexaShell: A Thin Harness for Exploring the Limits of VLM Agents as Robot Coding Agents","2026-10-16",["robotics","agent","code"],["software-engineering","robotics"],"hexafuture"],["2610.23374","MOSS-Image Technical Report","2026-10-16",["visual-gen"],["computer-science"],"openmoss"],["2610.23279","Xiaomi-Lens-0 Technical Report","2026-10-16",["vision","multimodal"],["computer-science"],"seerray-lab"],["2610.23230","RAP: A Dataset for Paper-to-Poster Generation via Declarative Markup from Human-Crafted Exemplars","2026-10-16",["benchmark","code"],["computer-science"],"meituan"],["2610.23181","Tree-Aware Posterior Distillation for Long-Horizon Agentic Reinforcement Learning","2026-10-16",["rl","agent"],["computer-science"],"tencent-hunyuan"],["2610.23121","Scene-Designer: Learning Design Intelligence for Agentic 3D Scene Generation","2026-10-16",["agent","vision"],["computer-science"],"zju-omniai"],["2610.22791","Self-Evolving Coding Agents via Harness-Policy Optimization","2026-10-16",["code","agent","rl"],["software-engineering"],"microsoft-research"],["2610.22471","Has GPT-6-Astra Achieved Embodied Intelligence?","2026-10-15",["robotics","benchmark"],["robotics"],"sensenova"],["2610.22350","FDE-Gym: The Last Mile of AI Delivery","2026-10-15",["agent","benchmark"],["computer-science"],"deeplethe"],["2610.22347","Speculative Forcing for Efficient Autoregressive Video Generation","2026-10-15",["visual-gen","efficiency"],["film-animation"],"naver-ai-lab"],["2610.22341","3DCodeVerse: Exploring 3D Worlds as Code","2026-10-15",["code","visual-gen"],["software-engineering"],"university-of-southern-california"],["2610.22276","MJev-Spatial: System One Spatial Intelligence via Native Attention Matching","2026-10-15",["multimodal","efficiency"],["computer-science"],"mind-lab"],["2610.22264","SCOPE: Semantic Bounded Credit Optimization for Prompt Engineering in Text-to-image Generation","2026-10-15",["rl","visual-gen"],["computer-science"],"university-of-rochester"],["2610.22200","OpenGCUA: Toward a Fully Open Stack for Generalist Computer Use","2026-10-15",["agent","code"],["computer-science"],"tongyi-mai"],["2610.22194","AutoSFT: An Agentic System for Automated Data Research in Supervised Fine-Tuning","2026-10-15",["agent","llm"],["computer-science"],"tsinghua-nlp-group"],["2610.22174","Mosaic: Multi-Session Agentic Reinforcement Learning","2026-10-15",["rl","agent"],["computer-science"],"lark-lab-hkust-gz"],["2610.22169","Lexical Reference Frames: Interpreting LM Residual Streams in Native Semantics","2026-10-15",["safety","llm"],["computer-science"],null],["2610.22131","Intern-Decision: Efficient Multimodal Decision Making with Autoregressive Masked Language Modeling","2026-10-15",["multimodal","agent","efficiency"],["computer-science"],"intern-large-models"],["2610.21783","Learning to Think and Explore in 3D World","2026-10-15",["multimodal","rl","reasoning"],["computer-science"],"university-of-california-at-san-diego"],["2610.21767","From Research Goals to Scientific Closure: Evaluating Agents Through Evidence-Supported Physics Research","2026-10-15",["ai4science","agent","benchmark"],["physics"],"shanghai-ai-laboratory"],["2610.21601","Waypoint-OPD: Transferring RL-Induced Policy Shifts via Gated On-Policy Distillation","2026-10-15",["rl","efficiency"],["computer-science"],null],["2610.21459","Can Visual Generation Expand What Unified Multimodal Models Know?","2026-10-15",["multimodal"],["computer-science"],"z-ai"],["2610.21306","VLA-$Δ$: Learning Physical State Transition for Vision-Language-Action Models","2026-10-15",["robotics"],["robotics"],"zju-omniai"],["2610.21261","Behavioral Compositionality of Sparse Autoencoder Representations","2026-10-15",["safety"],["computer-science"],"opensae"],["2610.21046","PremiseCheck: Efficient Fact-Checking of Human and AI Peer Reviews","2026-10-15",["benchmark","efficiency"],["computer-science"],null],["2610.23986","BlenderWorld: Benchmarking Visual Agents for 3D Creation from Real-World References","2026-10-14",["benchmark","agent","multimodal"],["materials","robotics"],"qwen"],["2610.20951","LEV: Thinking Fast and Deep with Looped Decision Models","2026-10-14",["reasoning","efficiency"],["computer-science"],null],["2610.20939","MAI-Gym: An Open All-in-One Platform for GUI Agent Evaluation and RL Training across Mobile and Computer Environments","2026-10-14",["agent","rl","benchmark"],["mathematics"],"tongyi-mai"],["2610.20818","Intern Lumina U2: A Multi-Codebook Diffusion Large Language Model for Omni-Visual Understanding and Image Generation","2026-10-14",["multimodal","visual-gen"],["computer-science"],"shanghai-ai-laboratory"],["2610.20693","Orbia: Do Generated Videos Form a Consistent 3D World?","2026-10-14",["visual-gen","benchmark"],["computer-science"],"tencent"],["2610.20584","Player-1: Learning to Play from Synthetic Experience","2026-10-14",["agent","rl"],["computer-science"],"arc-lab-tencent"],["2610.20572","Recovering Reliable Supervision via Critique-Guided On-Policy Distillation","2026-10-14",["efficiency","reasoning"],["mathematics"],null],["2610.20571","Any Time, Any Space: A World-State-Centric Framework for Agentic Video Generation","2026-10-14",["visual-gen","agent"],["film-animation"],"zhejiang-university"],["2610.20556","Speculative Sampling in Interactive Video Generation","2026-10-14",["visual-gen","efficiency"],["film-animation"],"university-of-science-and-technology-of-china"],["2610.20442","Towards Agentic Robots: Model, Data, Environment, and Harness","2026-10-14",["robotics","agent"],["robotics"],null],["2610.20424","BizAgentBench: Unveiling Hidden Challenges in Deploying Long-Horizon Agents for Real-World Business","2026-10-14",["agent","benchmark"],["finance"],"ant-international"],["2610.20399","Cortex: Consolidating Context into Parametric Memory","2026-10-14",["llm","efficiency"],["life-sciences"],"bytedance-seed"],["2610.20344","Doc2Harness: Benchmarking LLMs on Long-Horizon Harness Engineering of Coding Agents","2026-10-14",["agent","benchmark","code"],["software-engineering"],"institute-of-computing-technology"],["2610.20341","InfiniSplat v2: Analyzing and Improving Feed-Forward 3DGS from Any Views","2026-10-14",["visual-gen","vision"],["computer-science"],"zju3dv"],["2610.19975","LatentLoop: Scaling Looped MoE Models with Asymmetric Compute Allocation","2026-10-14",["efficiency"],["computer-science"],null],["2610.19849","DeltaStreamer: Towards Human-Like Agentic Streaming Video Understanding","2026-10-14",["multimodal","agent"],["computer-science"],"bytedance"],["2610.19820","ActionOPD: Distilling Decisions, Not Tokens","2026-10-14",["agent","efficiency"],["computer-science"],"tsinghua-nics-efc"],["2610.19789","More Methods, Better Agents? A Reality Check on Self-Distillation in Agentic Reinforcement Learning","2026-10-14",["agent","rl","benchmark"],["computer-science"],"zhejiang-university"],["2610.19686","Interlock: From Local Correctness to Task Completion in Long-Horizon Agents","2026-10-14",["agent","benchmark"],["computer-science"],"longcat"],["2610.19366","Seeing, Maintaining, and Learning: The Evolution of Multimodal Agent Memory","2026-10-14",["multimodal","agent"],["computer-science"],"fudan-university"],["2610.19159","Into the Unknown: Benchmarking AI agents on open scientific research problems","2026-10-13",["agent","ai4science","benchmark"],["chemistry","materials","physics"],"carnegie-mellon-university"],["2610.18861","Hybrid Reasoning Agents","2026-10-13",["reasoning","agent"],["computer-science"],"stanford-university"],["2610.18860","AesCode: Aesthetic Code Generation with Decoupled Cross-Modal Rewards","2026-10-13",["code","visual-gen","rl"],["software-engineering"],"microsoft"],["2610.18829","OmniCompass: A Holistic Benchmark for Omni-Modal Video Grounding and Understanding","2026-10-13",["multimodal","benchmark"],["computer-science"],"vclab"],["2610.18813","Self-Evolving Agent Harness","2026-10-13",["agent"],["computer-science"],"zgcagi"],["2610.18784","Pack Light, CarryOn: Building Object-Oriented Context for Long-Horizon Agents","2026-10-13",["agent","efficiency"],["computer-science"],"mcauley-lab"],["2610.18774","GreenWAM: A World Foundation Model for Robotics","2026-10-13",["robotics"],["robotics"],"sber-robotics-center"],["2610.18773","Jeva: System One Models from First Principles","2026-10-13",["llm","efficiency"],["computer-science"],"national-university-of-singapore"],["2610.18615","VideoJev: System One Model for Video Understanding and Interaction","2026-10-13",["multimodal","efficiency"],["computer-science"],"multimedia-computing-group-nanjing-university"],["2610.18356","Urbex: Agentic Spatial Grounding for City-Scale 3D Scenes","2026-10-13",["agent","vision","rl"],["computer-science"],"alibaba"],["2610.18072","LensErase: A Foundation Model for Erase Anything","2026-10-13",["visual-gen"],["computer-science"],"microsoft"],["2610.18067","WorkRover: Scaling Long-Horizon Cowork Agent Training Via Recursive Environment Evolution","2026-10-13",["agent","rl"],["computer-science"],"nex-agi"],["2610.17953","Agent-Reflex: Teaching Language Agents to Act Reflectively","2026-10-13",["agent","rl"],["computer-science"],"nlpir-lab-ruc"],["2610.17771","Many Right Answers, One Training Example: Removing Responder Noise from GRPO for LLM Recommenders","2026-10-13",["rl","llm"],["computer-science"],"nanyang-technological-university"],["2610.17665","RoTA-Bench: Evaluating 3D Spatial Reasoning through Interactive Projection Alignment","2026-10-13",["multimodal","benchmark","reasoning"],["computer-science"],"mac-automl"],["2610.17656","Open-o3-Video-v2: Seek, Ground, and Reason with Verifiable Spatio-Temporal Evidence","2026-10-13",["multimodal","reasoning","agent"],["computer-science"],"peking-university"],["2610.17552","FlashVSR+: Efficient Video Super-Resolution via Hybrid Attention and Reward-Yield Distribution Matching","2026-10-13",["visual-gen","efficiency"],["computer-science"],"mmlab-cuhk"],["2610.17309","StartLux-Decision: AutoThink Decision Models with Reinforcement Learning for Calibrated Deliberative Decisions","2026-10-12",["reasoning","rl"],["computer-science"],"startlux"],["2610.17227","AReA: Adaptive Receptive-Field Attention for Efficient High-Resolution Diffusion Transformers","2026-10-12",["visual-gen","efficiency"],["computer-science"],"university-of-toronto-computer-science"],["2610.17201","Multics: Generative Multi-Player World Model with Hierarchical Communication","2026-10-12",["visual-gen"],["film-animation"],"arc-lab-tencent"],["2610.16811","X-Ragent: Tool-Augmented Reinforcement Learning for Evidence-Grounded Chest X-Ray Reasoning","2026-10-12",["rl","reasoning","ai4science"],["clinical-medicine"],"carnegie-mellon-university"],["2610.16794","TPPO: Tracking-Value PPO For Language Model Reasoning","2026-10-12",["rl","reasoning"],["mathematics"],null],["2610.14993","Learning Where to Split: Dynamic Band Splitting for Audio Source Separation","2026-10-10",["multimodal"],["computer-science"],"tsinghua-machine-learning-group"],["2610.14766","Last Secure Code Benchmark: Is the Code You Vibe Secure?","2026-10-10",["code","safety","benchmark"],["software-engineering","cybersecurity"],"qwen"],["2610.14746","BridgeWAM: Bridging Video and Action Experts via Latent Bridge Queries","2026-10-10",["robotics","multimodal"],["robotics"],"fudan-university"],["2610.14709","Vela: User Behavior as Supervision Sparks LLM Intelligence","2026-10-10",["llm"],["computer-science"],"allspark-research"],["2610.14224","Harness^2: Recursive Agent Harnessing for an Open World","2026-10-09",["agent"],["computer-science"],"google"],["2610.14157","OpenMobile-2: Building Versatile Mobile Agents with Scalable Environments and App-Native Tools","2026-10-09",["agent","rl"],["computer-science"],"nanjing-university"],["2610.14136","Correction Training for Diffusion Models","2026-10-09",["visual-gen"],["computer-science"],"university-of-california-berkeley"],["2610.13912","RSI-Anything: Unlocking Recursive Self-Improvement with World Models","2026-10-09",["agent","ai4science"],["computer-science"],"phai-labs"],["2610.13630","Multi-Stream Attention Residuals","2026-10-09",["llm","efficiency"],["computer-science"],"shanghai-jiao-tong-university"],["2610.13623","APEX-Harm: Evaluating Agent Safety in Shared Professional Workspaces","2026-10-09",["safety","agent","benchmark"],["cybersecurity"],"king-abdullah-university-of-science-and-technology"],["2610.13527","S-Space: Exploring Spatial Workspace in Multimodal Language Models","2026-10-09",["multimodal","safety"],["computer-science"],"mirros"],["2610.13521","Beyond Domain-Routed Supervision: Cross-Teacher Guarding for Multi-Teacher On-Policy Distillation","2026-10-09",["efficiency"],["computer-science"],"jingdong"],["2610.13361","AgentSWE: Can Coding Agents Build the Agent You Actually Want?","2026-10-09",["agent","code","benchmark"],["software-engineering"],null],["2610.13166","Scaling Agentic Data for Long-Horizon Terminal Intelligence","2026-10-09",["agent","benchmark"],["computer-science"],"alibaba"],["2610.13021","EmbodiedCLBench: Evaluating Continual Learning for Self-Evolving Embodied Agents","2026-10-09",["robotics","agent","benchmark"],["robotics"],"pku-value-lab"],["2610.12984","Work2Skill: Capturing Human GUI Experience for Reusable Agent Skill Induction","2026-10-09",["agent","multimodal"],["computer-science"],"nex-agi"],["2610.12978","Learning Representations to Guide Pixel Diffusion Transformers","2026-10-09",["visual-gen"],["computer-science"],"alibaba"],["2610.12818","LeWAM: End-to-End World and Action Modeling with JEPAs","2026-10-08",["robotics","vision"],["computer-science"],"lewam"],["2610.12772","Reflex: Visually Grounded Reactive Humanoid Control","2026-10-08",["robotics","rl"],["robotics"],"university-of-washington"],["2610.12611","SoftwareGym: Scaling Computer-Use Agent Training with Diverse Synthesized Software Environments","2026-10-08",["agent","benchmark"],["computer-science"],"shanghai-jiao-tong-university"],["2610.12572","DiffusionART: Efficient, Stable and Scalable Reinforcement of Diffusion Models","2026-10-08",["visual-gen","rl"],["computer-science"],"klingai-research"],["2610.12461","OuroWorld: Bringing Any 3D World Alive as Diverse, Endlessly Looping 3D Cinemagraphs","2026-10-08",["visual-gen"],["film-animation"],"alaya-lab"],["2610.12374","AgentGarten: Code Worlds for Evolving Agents","2026-10-08",["agent","visual-gen"],["computer-science"],"mirros"],["2610.12299","Multi-Agent Egocentric World Model with Fine-Grained Embodied Interaction","2026-10-08",["visual-gen","robotics"],["robotics"],"kaist-ai"],["2610.12242","TokenRouter: Efficient Serving System for Token-Level LLM Routing","2026-10-08",["efficiency"],["computer-science"],"tsinghua-nics-efc"],["2610.12126","SuperNav: An Agentic Navigation System for Any Task in Any Scene","2026-10-08",["robotics","agent"],["robotics"],"zju3dv"],["2610.11959","MiMo-V2.6: Scaling Reinforcement Learning Towards Self-Improvement","2026-10-08",["llm","rl","multimodal"],["computer-science"],"xiaomi-mimo"],["2610.11765","Thinking Inertia: LLMs Keep Thinking When Told Not To","2026-10-08",["reasoning","llm"],["computer-science"],"stanford-university"],["2610.11287","REMORY: Learning Residual Memory for Context Compaction","2026-10-08",["agent","efficiency"],["computer-science"],"fusion-lab-generative-vision-lab-of-fudan-university-and-sii"],["2610.10539","Tetris3D: 3D Scene Generation With Objects That Fit Together","2026-10-07",["vision"],["physics"],"kaist-ai"],["2610.10528","Long-WAM: Scaling the Context of World-Action Models","2026-10-07",["robotics","efficiency"],["robotics"],"nvidia"],["2610.10524","GRACE: Generation-aware latent compression for efficient video generation","2026-10-07",["visual-gen","efficiency"],["film-animation"],"kaist-ai"],["2610.10429","SGF+: Decoupling Gradient Flows for Autoregressive Video Generation","2026-10-07",["visual-gen"],["film-animation"],null],["2610.10409","RobotWorld: Benchmarking Multimodal Agents for Robot Use Across Diverse Tasks and Embodiments","2026-10-07",["robotics","agent","benchmark"],["robotics"],null],["2610.09872","LiveMACE: Process-Aware Evaluation of LLM Agent Capabilities in Evolving Markets","2026-10-07",["agent","benchmark"],["finance","economics"],"fudan-university"],["2610.09823","UltraText Bench: A Comprehensive Bilingual Benchmark for Evaluating Visual Text Rendering in Image Generation","2026-10-07",["visual-gen","benchmark"],["computer-science"],"westlake-university"],["2610.09455","RLHND: Video Foundation Models as Physically Grounded Hand Trackers for Robot Learning","2026-10-07",["robotics","vision"],["robotics"],"rlwrld"],["2610.09450","Iris-3B: Going Beyond the Latent with Pixel-Space Diffusion Training, Conversion and Fine-Tuning","2026-10-07",["visual-gen"],["computer-science"],"speridlabs"],["2610.09326","VIS-Ground: Video Interactive Storytelling with Contextual Grounding","2026-10-07",["visual-gen"],["film-animation"],"google"],["2610.09221","Consistent Distribution Matching for Data-Free Diffusion Distillation","2026-10-06",["visual-gen","efficiency"],["computer-science"],"vector-institute"],["2610.09087","U-Space: Uncovering When and Why Uncertainty Arises in Language Models","2026-10-06",["safety","llm"],["computer-science"],"multimodal-ai-lab-technical-university-of-darmstadt"],["2610.08963","On KL-Regularized Policy Optimization","2026-10-06",["rl","llm"],["computer-science"],"math-ai"],["2610.08760","WorldSonus: Bringing Sound to Worlds","2026-10-06",["visual-gen","multimodal"],["computer-science"],"noizai"],["2610.08699","nanoMuse: An Open-Source Personal Agent for Every Device You Own","2026-10-06",["agent"],["computer-science"],"zhejiang-university"],["2610.08621","Recursive Game Creator: An Agentic Product-Level Experience-Oriented Game Harness","2026-10-06",["agent","code"],["computer-science"],"the-university-of-hong-kong"],["2610.08430","NeMo-DCR: Bit-Exact Delta-Compressed Refit for Scalable Agentic RL at Trillion-Parameter Scale","2026-10-06",["efficiency","rl"],["computer-science"],"nvidia"],["2610.08215","Learn2Play Bench: How Well Do LLM Agents Learn from Experience in Unfamiliar Environments?","2026-10-06",["agent","benchmark"],["computer-science"],"national-university-of-singapore"],["2610.08077","Self-Retrospection Distillation: Turning Post-hoc Experiences into Prior Foresight","2026-10-06",["rl","agent"],["computer-science"],"salesforce-ai-research"],["2610.07987","VisionWeave: Weaving Elastic Visual Representations as a Native Capability of MLLMs","2026-10-06",["multimodal","efficiency"],["computer-science"],"qwen"],["2610.07967","DecepEval: A Benchmark for Evaluating Deception in LLM Agents","2026-10-06",["safety","benchmark","agent"],["computer-science"],"xi-an-jiaotong-university"],["2610.07767","TRACE: Rollout-Guided Quantization-Aware Training for FP4 Reinforcement Learning of MoE Language Models","2026-10-06",["efficiency","rl","llm"],["computer-science"],"qwen"],["2610.07625","Stateless Language Agents: Scaling Long-Horizon Automated Research","2026-10-06",["agent","ai4science"],["computer-science"],"stanford-university"],["2610.07591","Recurrent Looped Transformer","2026-10-06",["llm","reasoning"],["computer-science"],"princeton-university"],["2610.07557","CheckerBench: Can Long-Horizon Agents Synthesize Static-Analysis Checkers?","2026-10-06",["code","agent","benchmark"],["cybersecurity"],"humanlaya-data-lab"],["2610.06833","Towards Looped Models Done Right, Part II: Rethinking at Fixed Points","2026-10-05",["efficiency","llm"],["cybersecurity"],"institute-of-foundation-models"],["2610.06830","MemPilot: Orchestrating On-Demand Multimodal Memory Curation for LLM Agents","2026-10-05",["agent","multimodal"],["computer-science"],"nanyang-technological-university"],["2610.06801","MC-Sparse: Deconstructing and Closing the Dense-Sparse Attention Gap in Diffusion Transformers","2026-10-05",["visual-gen","efficiency"],["computer-science"],"tencent-hunyuan"],["2610.06647","LoGRA: Scaling LLM Reinforcement Learning with Low-Rank Gradient Sketches","2026-10-05",["rl","efficiency"],["computer-science"],"nvidia"],["2610.06617","RealtimeWAM: One-Step Asynchronous World Action Models","2026-10-05",["robotics","efficiency"],["robotics","film-animation"],"nanyang-technological-university"],["2610.06577","Optimizing the Optimizer: Language Models Discover Faster Molecular Relaxation Algorithms","2026-10-05",["ai4science","agent"],["chemistry","physics"],null],["2610.06293","VepAgent: Bridging Causal-Transition via Tool-Augmented Reinforcement Learning for Video Event Prediction","2026-10-05",["multimodal","rl","reasoning"],["computer-science"],"ucsc-vlaa"],["2610.06100","From Traces to Agentic Worlds: Agentic Language World Models for Interactive Environment Simulation","2026-10-05",["agent"],["computer-science"],"nanyang-technological-university-singapore"],["2610.05608","Kandinsky 6.0 Video: Foundation Models for Synchronized Video and Audio Generation","2026-10-04",["visual-gen","multimodal"],["film-animation"],"kandinsky-lab"],["2610.05416","Prism: Dynamic Sparse Attention for Native 2K Joint Video-Audio Generation Model Training","2026-10-04",["visual-gen","efficiency"],["computer-science"],"tencent-hunyuan"],["2610.05162","Memadapter: Counterfactual Adaptation Against Memory-induced Sycophancy","2026-10-04",["alignment","agent"],["computer-science"],"deep-group-at-jilin-university"],["2610.05107","SearchJev: A Fast and Calibrated System-1 Model for Search Agents","2026-10-04",["agent","efficiency"],["computer-science"],null],["2610.05030","EVISKILL: Grounding Skill Evolution in Replayable Evidence","2026-10-04",["agent"],["computer-science"],"deep-group-at-jilin-university"],["2610.04781","Super-Resolution in The Right Latent Space: A Frozen Vision-Foundation Substrate","2026-10-03",["vision","visual-gen"],["computer-science"],"rtp-llm"],["2610.04299","Questioning the Questions: Sustaining Self-Evolution in Reasoning Models","2026-10-03",["reasoning","rl"],["mathematics"],"huang-s-intelligence-lab"],["2610.04198","ALoDLM: Adaptively Looped Diffusion Language Models","2026-10-03",["llm","efficiency"],["computer-science"],"amazon"],["2610.03715","4DCodeBench: Benchmarking Agents on Inverse Graphics of Dynamic Scenes","2026-10-02",["benchmark","code","vision"],["software-engineering"],"4dcodebench"],["2610.03543","DuoMatching: Joint-Marginal Distribution Matching for Few-Step Video Generation","2026-10-02",["visual-gen","efficiency"],["film-animation"],"bytedance"],["2610.03391","Native Action-Prior Learning from Videos for World Action Models","2026-10-02",["robotics"],["robotics"],"meta"],["2610.03195","Source Preference in the Wild: How LLM Agents Favor Items by Source, and How to Reduce It","2026-10-02",["alignment","agent"],["computer-science"],"seoul-national-university"],["2610.02826","Scaling Trajectories for Complex Tasks through Recursive Self-Rewrite","2026-10-02",["agent","rl"],["computer-science"],"tencent-hunyuan"],["2610.02521","Spatial Memory Intelligence: Endowing World Models with Understanding-Driven Long-Term Memory","2026-10-01",["visual-gen","multimodal"],["robotics","film-animation"],null],["2610.02508","World Action Modeling with Progressive Visual Planning","2026-10-01",["robotics","efficiency"],["robotics"],"meta"],["2610.02368","Rethinking World-Action Model for Compositional and In-Context Robotic Manipulation","2026-10-01",["robotics","reasoning"],["robotics"],"dagroup-pku"],["2610.02304","SimuVerity: Benchmarking Agents for Engineering-Grade Simulink Model Generation","2026-10-01",["agent","benchmark","code"],["computer-science"],"xi-an-jiaotong-university"],["2610.02298","EditHero: A Benchmark for Long-Horizon Part-Level 3D Editing and Vibe Modeling","2026-10-01",["vision","benchmark"],["computer-science"],null],["2610.02205","PROWBench: Do Video Models Render What the Program Specifies?","2026-10-01",["visual-gen","benchmark"],["computer-science"],"alaya-lab"],["2610.02193","Hierarchical Continuous Diffusion Language Models","2026-10-01",["llm"],["computer-science"],"university-of-illinois-at-urbana-champaign"],["2610.02162","World Observer: Joint Actor-Observer Generation for Persistent World Modeling","2026-10-01",["visual-gen"],["computer-science"],"kaist-ai"],["2610.02160","4Director: Controlling Video World Models with Rigid 3D Geometry","2026-10-01",["visual-gen"],["computer-science"],"stability-ai"],["2610.02054","UniWAM: Unified World-Action Model","2026-10-01",["robotics","multimodal"],["robotics","film-animation"],"hkustgz"],["2610.01939","Fewer Tokens, Better Action: GPT-6 Astra Robot Agents with 14% Higher Success Rate but 65% Fewer Tokens","2026-10-01",["robotics","agent","efficiency"],["mathematics","robotics"],"dagroup-pku"],["2610.01892","Selection-Based Structured Reasoning: Toward Efficient Multimodal Search Agents","2026-10-01",["reasoning","efficiency","agent"],["computer-science"],"carnegie-mellon-university"],["2610.01780","RealCompanion: Benchmarking Human Understanding from Reasoning over Longitudinal Real-World Conversations","2026-10-01",["benchmark","agent"],["computer-science"],"quis-lab"],["2610.01762","OneStreamer: Unifying Perception, Memory, and Proactive Response in Streaming Video Interaction","2026-10-01",["multimodal"],["computer-science"],"nanjing-university"],["2610.01509","Sharpening Tax in Post-Training","2026-10-01",["rl","agent"],["mathematics"],"meta"],["2610.01496","SALD: Self-Referenced Advantage Learning for Diffusion Models","2026-10-01",["visual-gen"],["computer-science"],"indian-institute-of-technology-kanpur"],["2610.01415","Beyond Memory: Harnessing Long-Horizon Agents with Explicit Belief States","2026-10-01",["agent"],["computer-science"],"alibaba"],["2610.01257","Science Utopia? Closed-Loop LLM Simulation of Academic Research Ecosystems","2026-10-01",["agent","ai4science"],["economics"],"georgia-institute-of-technology"],["2610.01215","AutoGUIWorld: Image Generators as Visual World Models for GUI Agent","2026-10-01",["agent","visual-gen"],["computer-science"],"tencent-hunyuan"],["2610.00972","VeriHarness: Scaling Agentic Verification for Long-Horizon Tasks","2026-10-01",["agent","reasoning"],["computer-science"],"google"],["2610.00906","ActiveSaddler: Automated Curriculum Learning for Agent Harness Optimization","2026-10-01",["agent"],["computer-science"],"microsoft"],["2610.00812","Video Generation Models: A Survey of Post-Training and Alignment","2026-09-30",["visual-gen","alignment"],["film-animation"],"arizona-state-university"],["2610.00559","PhysVista: Benchmarking Physical Intelligence in VLMs via a Perception-Reasoning-Assessment Loop","2026-09-30",["multimodal","benchmark","reasoning"],["computer-science"],"university-of-science-and-technology-of-china"],["2610.00531","Science or Slop?: Benchmarking and Mitigating Scientific Slop in AI-Generated Papers","2026-09-30",["benchmark","ai4science"],["computer-science"],"seoul-national-university"],["2609.40340","EvoDuet: Bilevel Co-Evolution of Web Searching and Task Solving for Scientific Discovery","2026-09-30",["ai4science","agent"],["computer-science"],"minnesota-nlp"],["2609.40325","WorldAuditBench: Interactive 3D World Auditing with Multimodal Agents","2026-09-30",["multimodal","agent","benchmark"],["computer-science"],"university-of-california-santa-barbara"],["2609.39982","Mid-Harness: Scaling Actions Between Model and Harness for Terminal Agents","2026-09-30",["agent","efficiency"],["computer-science"],"nvidia"],["2609.39601","GroundingPI: A Grounding Foundation Model towards Physical Intelligence with Visual Primitives","2026-09-30",["robotics","vision"],["robotics"],"xpeng-ai"],["2609.39600","GroundAnything: Reconciling Parallel Decoding with Precise Visual Grounding at Flash Speed","2026-09-30",["vision","efficiency"],["computer-science"],"xpeng-ai"],["2609.39378","EgoTools: Towards Tool-Centric Reasoning in Real-World Egocentric Videos","2026-09-30",["multimodal","reasoning","benchmark"],["robotics"],"ropedia"],["2609.39102","False Frontiers: Diagnosing and Mitigating Co-Cheating in Self-Evolving Search Agents","2026-09-30",["agent","rl"],["computer-science"],"rutgers-university"],["2609.39045","RSIGame: Autonomous Agentic Game Development with Recursive Self-improvement","2026-09-30",["agent","code"],["computer-science"],"rsigame"],["2609.38923","GraphForge: Training Working Agents with Graph-Anchored Workspace Synthesis","2026-09-30",["agent","benchmark"],["computer-science"],"university-of-science-and-technology-of-china"],["2609.38879","Does Learning Protein Folding Generalize to Broader Reasoning?","2026-09-30",["ai4science","reasoning"],["life-sciences"],"shanghai-jiaotong-university"],["2609.38847","Scoring Higher, Answering Worse: Mitigating Reward Hacking in Rubric-Based RL via Protocol-Level Rubrics","2026-09-30",["rl","alignment","ai4science"],["clinical-medicine"],"bytedance"],["2609.38839","FrameMorrow: Future-guided Frame Selection with Prospective Tokens for Long-Horizon Video Generation","2026-09-30",["visual-gen","efficiency"],["film-animation"],"national-university-of-singapore"],["2609.38827","More Choices, Fewer Decisions: Ordinal-Scale Bias in JEV-like Direct-Decision Models","2026-09-30",["benchmark"],["computer-science"],"jilin-university"],["2609.38748","Hear the World in Stereo: Learning Dynamic Spatial Correspondence for Immersive Joint Video-Audio Generation","2026-09-30",["multimodal","visual-gen"],["computer-science"],null],["2609.38721","UniEvo-VL: An On-policy Self-Distillation Training Recipe for Multimodal Model Self-improvement","2026-09-30",["multimodal"],["computer-science"],"stanford-nlp"],["2610.00314","Predictive Credit: Measuring What Scientific Explanations Add to Experimental Forecasts","2026-09-29",["ai4science","benchmark"],["computer-science"],"carnegie-mellon-university"],["2609.38653","TERRA: Terrain-Aware Reconstruction, Retargeting and Control for Musculoskeletal Locomotion","2026-09-29",["robotics","rl"],["computer-science"],"mathis-group-epfl"],["2609.38597","PixelUMM: Encoder-Free Unified Image and Video Understanding and Generation","2026-09-29",["multimodal","visual-gen"],["computer-science"],"nvidia"],["2609.38537","Systematically Exploring the Capabilities of GPT-6 Astra as Embodied Policies","2026-09-29",["robotics","benchmark"],["robotics"],"galbot"],["2609.38445","AIM: Agentic Idea Management for Automated Research","2026-09-29",["agent","ai4science"],["computer-science"],"google"],["2609.38426","LoopVL: Recurrent Visual Intelligence","2026-09-29",["multimodal","reasoning"],["computer-science"],"renmin-university-of-china"],["2609.38334","EVOKE: Eliciting World Knowledge in Agents for Transferable Decision-Making","2026-09-29",["agent"],["computer-science"],"shanghai-jiao-tong-university"],["2609.38288","AREX-2: Advancing Self-Improving Agents through Long-Horizon Reflective Tasks","2026-09-29",["agent","rl"],["software-engineering"],"beijing-academy-of-artificial-intelligence"],["2609.38177","Imagine3D-LLM: Teaching MLLMs to Imagine 3D Scenes Before Answering","2026-09-29",["multimodal","reasoning"],["computer-science"],"kaist-ai"],["2609.38173","In-context Robot Learning Made Simple: A Democratized Recipe for Manipulation Tasks","2026-09-29",["robotics"],["robotics"],null],["2609.38169","STEPQuant: When and Where Errors Matter in Delta-Rule Recurrent State Quantization","2026-09-29",["efficiency"],["computer-science"],"zhejiang-university"],["2609.38155","Beyond the Timeline: Augmenting Long-Video Memory with Grounded Entity Biographies","2026-09-29",["multimodal"],["computer-science"],"amazon-science"],["2609.38143","Learning Meta-Skills for Agent Harness Design in Test-Time AI4AI","2026-09-29",["agent"],["finance"],"apodex"],["2609.38123","HelixWorld: A Real-time Interactive Audio-Visual World Model","2026-09-29",["visual-gen","multimodal"],["computer-science"],"noizai"],["2609.38079","OmniTaskonomy: When Does Visual Generation Improve Visual Understanding?","2026-09-29",["multimodal","vision"],["computer-science"],"uc-berkeley"],["2609.38078","MotorMind: Scaffolding General Vision Language Models for Zero-Shot Robot Manipulation","2026-09-29",["robotics","multimodal"],["software-engineering","robotics"],"university-of-illinois-at-urbana-champaign"],["2609.38057","EVO-WAM: Evolving World Action Models through Video-Action Verification","2026-09-29",["robotics"],["robotics"],"harbin-institute-of-techonology-shenzhen"],["2609.37959","TabFM: A Zero-Shot Foundation Model for Tabular Data","2026-09-29",["llm","benchmark"],["computer-science"],"google"],["2609.37725","Context Language Models","2026-09-29",["llm","agent"],["computer-science"],"meta"],["2609.37686","EngiWorld: What Can Frontier Agents Deliver in Professional Engineering Environments?","2026-09-29",["agent","benchmark"],["computer-science"],"zhiman-ai"],["2609.37533","E-MoE: Enhanced Mixture-of-Experts for Non-Factorized Diffusion Language Models","2026-09-29",["llm","efficiency"],["computer-science"],null],["2609.37372","Think Before You Score: Thinking Reward Model for Visual Generation","2026-09-29",["visual-gen","alignment"],["computer-science"],null],["2609.37226","Follow the Entities: A Corpus Map for Agentic Search","2026-09-29",["agent","llm"],["computer-science"],"microsoft"],["2609.37200","Adaptive Reward Routing: Dynamic Multi-Reward Optimization for Joint Audio-Video Diffusion via Forward-Process RL","2026-09-29",["visual-gen","rl"],["computer-science"],"tencent"],["2609.36601","SAKI: Maximal-Coupling-Routed Teacher Supervision for On-Policy Distillation","2026-09-29",["efficiency"],["computer-science"],"meituan"],["2609.36484","The Teacher Is a Direction, Not a Destination: Extrapolating RL-Induced Representation Residuals in On-Policy Distillation","2026-09-29",["efficiency","rl"],["computer-science"],null],["2609.36380","LEGO-Anything: Coding Agents for 3D Scene Reconstruction","2026-09-28",["code","vision","agent"],["software-engineering"],"amazon-web-services"],["2609.36322","Periodic Weak Spots: Phase Sensitivity from Chunked KV-Cache Compression","2026-09-28",["efficiency"],["computer-science"],"bytedance-seed"],["2609.36071","LongCat-DeepResearch Technical Report","2026-09-28",["agent"],["computer-science"],"longcat"],["2609.36012","In-Context Learning for Robots: Methods and Applications","2026-09-28",["robotics"],["robotics"],"knowin-ai"],["2609.35767","Learning Native Reflection in Unified Models with Interleaved Reinforcement Learning","2026-09-28",["multimodal","rl","visual-gen"],["computer-science"],"mmlab-ntu"],["2609.35748","Improving Test-Time Scaling with Adaptive Looped Transformers","2026-09-28",["reasoning","efficiency"],["computer-science"],"tsinghua-nics-efc"],["2609.35690","Agent Priors-guided Policy Learning","2026-09-28",["robotics","agent"],["robotics"],"national-university-of-singapore"],["2609.35432","Self-Evolving Coding Agents: From Digital Programs to Physical-World Intelligence","2026-09-28",["robotics","code","agent"],["software-engineering","robotics"],"hexafuture"],["2609.35427","LLMs are General Asynchronous Agents","2026-09-28",["agent"],["robotics"],"yandex-research"],["2609.35347","Beyond Teacher Assignment: Domain-Normalized Multi-Teacher On-Policy Distillation","2026-09-28",["efficiency","rl"],["mathematics"],"nanyang-technological-university"],["2609.35259","On-Policy or Off-Policy Learning? A Systematic Study of Distillation Dynamics","2026-09-28",["efficiency","reasoning"],["clinical-medicine"],"university-of-cambridge"],["2609.34981","What Makes World Action Models Generalize? An Empirical Study of Test-Time Future Modeling","2026-09-28",["robotics"],["robotics"],"tsinghua-leaplab"],["2609.34759","PanoVLN: Towards Effective Panoramic Vision-and-Language Navigation","2026-09-28",["robotics","multimodal"],["computer-science"],"zhejiang-university"],["2609.34715","PDE-JEPA: Predictive Representation Learning of Latent Dynamics Modeling for Parametric PDEs","2026-09-28",["ai4science","efficiency"],["computer-science"],"zhejiang-university"],["2609.34563","Rethinking Latent Visual Reasoning: Grounding Latent Reasoning in Visual Evidence","2026-09-28",["reasoning","multimodal"],["computer-science"],"amazon"],["2609.34309","MaLiang-Harness: A Programmable Path to Image and Video Generation","2026-09-28",["visual-gen","code"],["film-animation"],"national-university-of-singapore"],["2609.33803","Diffusion Reward Models","2026-09-27",["alignment","rl"],["computer-science"],"tsinghua-nlp-group"],["2609.33757","YuE2: Unifying Symbolic and Audio Music Generation at Frontier Quality","2026-09-27",["multimodal","visual-gen"],["music"],"multimodal-art-projection"],["2609.33439","Raven: The Harness of Harnesses for Composable Agentic Intelligence","2026-09-27",["agent"],["computer-science"],"evermind"],["2609.33325","VisionHOPE: Visual Backbones as Self-Modifying Learning Systems","2026-09-27",["vision"],["computer-science"],"mininglamp-technology"],["2609.33304","Relevance Does Not Imply Applicability: Experience Activation for Personal GUI Agents","2026-09-27",["agent"],["computer-science"],"nanyang-technological-university"],["2609.33295","TraceDance: An Automated System for Building Agent Behavior Benchmarks from Real-World Agent Deployment Traces","2026-09-27",["agent","benchmark","safety"],["computer-science"],"bytedance"],["2610.13990","DoSA: Dots Sparse Attention","2026-09-26",["efficiency"],["computer-science"],"dots-studio"],["2609.32993","X-Tree: Tokenizing Reusable Experience for Efficient Agent Generalization","2026-09-26",["agent","efficiency"],["computer-science"],"university-of-waterloo"],["2609.32722","Scaling Properties of Same-Family On-Policy Distillation","2026-09-26",["efficiency","rl"],["mathematics"],"zhejiang-university"],["2609.32712","MassAlloc Attention: Let Attention Allocate Its Own Compute","2026-09-26",["efficiency"],["computer-science"],"beijing-academy-of-artificial-intelligence"],["2609.32704","CoWindow Attention: Full Causal Coverage Is a Collective Property","2026-09-26",["efficiency"],["computer-science"],"beijing-academy-of-artificial-intelligence"],["2609.32607","VoxMem: Benchmarking Multimodal Memory in Large Audio Language Models","2026-09-26",["multimodal","benchmark"],["computer-science"],"the-university-of-melbourne"],["2609.32577","Groupwise Agentic Grading and Advantage Redistribution for Code Agent RL","2026-09-26",["code","rl"],["computer-science"],"xiaomi-mimo"],["2609.32259","Prefill-Free Cross-Family KV Cache Transfer for Heterogeneous Multi-Agent LLMs","2026-09-26",["efficiency","agent"],["computer-science"],"university-of-southern-california"],["2609.32019","Decentralized Master-Mind: Joint Action Refinement through Iterative Intent Denoising in Multi-Agent Pathfinding","2026-09-25",["robotics","rl"],["computer-science"],null],["2609.31948","Duplex-MPE: Benchmarking Multi-Party Interaction in Full-Duplex Dialogue","2026-09-25",["multimodal","benchmark"],["computer-science"],"peking-university"],["2609.31847","Omni-IO Skills: Harnessing Your Agent Omni-Native","2026-09-25",["agent","multimodal"],["computer-science"],"national-university-of-singapore"],["2609.31620","FuseReg: Regularizing Layer Fusion Mitigates the Reconstruction-Generation Gap in Representation Autoencoders","2026-09-25",["visual-gen"],["computer-science"],"usc-physical-superintelligence-psi-lab"],["2609.30467","Where Does Retrieval-Based Open-Ended Evaluation Fail? Automatic Taxonomy Induction from Long-Form Medical Answer Factuality Verification","2026-09-24",["ai4science","benchmark"],["clinical-medicine"],"center-for-language-and-speech-processing-jhu"],["2609.30199","ExplorationBench: Measuring AI Systems' Exploration in Verifiable Alien Worlds","2026-09-24",["ai4science","benchmark","agent"],["computer-science"],"tencent-hunyuan"],["2609.29892","Qwen-Planner-Agent: A Closed-Loop AI-for-AI Framework for Real-World Mobile Planner Agents","2026-09-24",["agent","rl"],["computer-science"],"tongyi-mai"],["2609.29429","Just Ask Jev: Reinforcement Learning for Calibrated Decisions as a Zero-Shot Detector of AI Alignment Failures","2026-09-24",["safety","alignment"],["cybersecurity"],null],["2609.29233","Post-Training Leaves Behavioral Shadows on Unrelated Decisions","2026-09-24",["llm","safety"],["computer-science"],"peking-university"],["2609.29167","IndicBankBench: Evaluating Safety and Reliability of Language Model Assistants in Indian Retail Banking","2026-09-24",["safety","benchmark"],["finance"],"national-payments-corporation-of-india"],["2609.28654","Training Object Permanence in World Models","2026-09-23",["visual-gen","reasoning"],["film-animation"],"carnegie-mellon-university"],["2609.28466","The Past Frames the Future: Memory for Autoregressive Video Generation","2026-09-23",["visual-gen"],["film-animation"],null],["2609.26780","SpeakerMem-R1: Speaker-Centered Dual-Track Memory for Multi-Party Dialogue","2026-09-22",["agent","rl"],["computer-science"],"zhejiang-university"],["2609.26550","JEV-as-a-Judge: Accept When Confident, Escalate When Unsure","2026-09-22",["benchmark","efficiency"],["economics"],"carnegie-mellon-university"],["2609.26333","Disaggregated Quantization: Specializing LLM Prefill and Decode","2026-09-22",["efficiency"],["computer-science"],"ist-austria-distributed-algorithms-and-systems-lab"],["2609.25804","The Tasteful Agent: Measuring and Improving Taste in Long-Horizon Tasks","2026-09-22",["agent","benchmark"],["computer-science"],"microsoft"],["2609.25270","RULER: Instance-aware Rubric Rewards for SVG Generation","2026-09-21",["rl","visual-gen"],["computer-science"],"inclusionai"],["2609.25186","From Pattern Recognizers to Personalized Companions: A Survey of Large Language Models in Mental Health","2026-09-21",["llm","ai4science"],["clinical-medicine","psychology"],null],["2609.25165","Ovis-Embedding: Pushing the Frontiers of Universal Omni-Modal Embeddings","2026-09-21",["multimodal"],["computer-science"],"ath-maas"],["2609.25001","GameHorizon Suite: Multi-Horizon Data and Evaluation in Gameplay","2026-09-21",["benchmark","agent"],["computer-science"],"tencent"],["2609.24997","VideoGen-Agent: Reinforcing Video Generation Agents","2026-09-21",["visual-gen","agent","rl"],["film-animation"],"princeton-university"],["2609.24984","WorldCrafter: Consistent Video World Model with Implicit 3D-aware Memory","2026-09-21",["visual-gen"],["computer-science"],"arc-lab-tencent"],["2609.24983","onPanda: Efficient Annotation of On-Policy Alignment Data for LLMs and Agents via Token-Level Correction","2026-09-21",["alignment","agent"],["computer-science"],"stepfun"],["2609.24981","GAE: Learning a Geometry-Native Latent Space for 3D-Consistent World Generation","2026-09-21",["visual-gen","vision"],["computer-science"],"arc-lab-tencent"],["2609.24974","Harness-Zero: Harness Distillation via Agent-as-Harness","2026-09-21",["agent","efficiency"],["computer-science"],"peking-university"],["2609.24972","RRSI: Regularized Recursive Self-Improvement of Agent Harnesses","2026-09-21",["agent","benchmark"],["computer-science"],"google"],["2609.24058","All-in-One Multilingual Scene Text Recognition with Script-aware Mixture-of-Experts","2026-09-21",["vision","efficiency"],["computer-science"],"fudan-university"],["2609.23986","Jev-Mem: System-One-Controlled Agentic Memory for Efficient AI Agents","2026-09-21",["agent","efficiency"],["economics"],null],["2609.23863","Grounded Action Model: 3D Grounding as a Foundation for Robotics","2026-09-20",["robotics"],["robotics","film-animation"],"national-university-of-singapore"],["2609.23377","One to More, More to One: Category-Aware Iterative Expert Training for Software Engineering Agents","2026-09-20",["code","rl","agent"],["software-engineering"],"alibaba"],["2609.23088","OmniEdu: Open Foundation Models for Learning and Teaching","2026-09-19",["llm"],["education"],"peking-university"],["2609.23038","Spatial-Interactor: Learning Spatial Reasoning through Interaction with the Observable Physical World","2026-09-19",["multimodal","reasoning"],["computer-science"],"zju-omniai"],["2609.22966","Transferring the Intelligence of VLMs to Robotic Control","2026-09-19",["robotics","multimodal"],["robotics"],null],["2609.22068","CodeMidas: Scaling Agentic Coding RL Environments from Code Itself","2026-09-18",["code","rl","agent"],["software-engineering"],"xiaomi-mimo"],["2609.22000","RecreationWorld: Scalable and Verifiable Environments for Hybrid Computer-Use Agents","2026-09-18",["agent","code","benchmark"],["computer-science"],"qwen"],["2609.21465","OmniVChat: Synthesizing, Benchmarking, and Training for Native Audio-Visual Dialogue","2026-09-18",["multimodal","benchmark"],["computer-science"],"qwen"],["2609.18703","RayOrch: Programming and Executing Lineage-Controlled Multi-Grain Dataflows for Foundation-Model Data Preparation","2026-09-16",["efficiency"],["software-engineering"],"peking-university"],["2609.15779","EvoOntology: A Self-Evolving Ontology Layer for Data Agents","2026-09-14",["agent"],["computer-science"],"ruc-datalab"],["2609.13814","Realtime-Venus: A full-duplex interaction system with asynchronous delegation","2026-09-12",["multimodal","agent"],["computer-science"],"inclusionai"]]}