ずんだもんのHugging Faceニュース

【本日の論文】 1. WildDet3D: Scaling Promptable 3D Detection in the Wild https://huggingface.co/papers/2604.08626 2. FORGE:Fine-grained Multimodal Evaluation for Manufacturing Scenarios https://huggingface.co/papers/2604.07413 3. RefineAnything: Multimodal Region-Specific Refinement for Perfect Local Details https://huggingface.co/papers/2604.06870 4. EXAONE 4.5 Technical Report https://huggingfac...

【本日の論文】 1. Rethinking Generalization in Reasoning SFT: A Conditional Analysis on Optimization, Data, and Model Capability https://huggingface.co/papers/2604.06628 2. SkillClaw: Let Skills Evolve Collectively with Agentic Evolver https://huggingface.co/papers/2604.08377 3. ClawBench: Can AI Agents Complete Everyday Online Tasks? https://huggingface.co/papers/2604.08523 4. HY-Embodied-0.5: Em...

【本日の論文】 1. SkillClaw: Let Skills Evolve Collectively with Agentic Evolver https://huggingface.co/papers/2604.08377 2. Rethinking Generalization in Reasoning SFT: A Conditional Analysis on Optimization, Data, and Model Capability https://huggingface.co/papers/2604.06628 3. HY-Embodied-0.5: Embodied Foundation Models for Real-World Agents https://huggingface.co/papers/2604.07430 4. When Numbe...

【本日の論文】 1. Rethinking Generalization in Reasoning SFT: A Conditional Analysis on Optimization, Data, and Model Capability https://huggingface.co/papers/2604.06628 2. SkillClaw: Let Skills Evolve Collectively with Agentic Evolver https://huggingface.co/papers/2604.08377 3. HY-Embodied-0.5: Embodied Foundation Models for Real-World Agents https://huggingface.co/papers/2604.07430 4. When Numbe...

【本日の論文】 1. Think in Strokes, Not Pixels: Process-Driven Image Generation via Interleaved Reasoning https://huggingface.co/papers/2604.04746 2. RAGEN-2: Reasoning Collapse in Agentic RL https://huggingface.co/papers/2604.06268 3. MARS: Enabling Autoregressive Models Multi-Token Generation https://huggingface.co/papers/2604.07023 4. Combee: Scaling Prompt Learning for Self-Improving Language...

【本日の論文】 1. Video-MME-v2: Towards the Next Stage in Benchmarks for Comprehensive Video Understanding https://huggingface.co/papers/2604.05015 2. Claw-Eval: Toward Trustworthy Evaluation of Autonomous Agents https://huggingface.co/papers/2604.06132 3. Learning to Retrieve from Agent Trajectories https://huggingface.co/papers/2604.04949 4. ACES: Who Tests the Tests? Leave-One-Out AUC Consisten...

【本日の論文】 1. OpenWorldLib: A Unified Codebase and Definition of Advanced World Models https://huggingface.co/papers/2604.04707 2. MinerU2.5-Pro: Pushing the Limits of Data-Centric Document Parsing at Scale https://huggingface.co/papers/2604.04771 3. LIBERO-Para: A Diagnostic Benchmark and Metrics for Paraphrase Robustness in VLA Models https://huggingface.co/papers/2603.28301 4. TriAttention:...

【本日の論文】 1. Self-Distilled RLVR https://huggingface.co/papers/2604.03128 2. A Simple Baseline for Streaming Video Understanding https://huggingface.co/papers/2604.02317 3. Token Warping Helps MLLMs Look from Nearby Viewpoints https://huggingface.co/papers/2604.02870 4. Agentic-MME: What Agentic Capability Really Brings to Multimodal Intelligence? https://huggingface.co/papers/2604.03016 5. T...

【本日の論文】 1. DataFlex: A Unified Framework for Data-Centric Dynamic Training of Large Language Models https://huggingface.co/papers/2603.26164 2. The Latent Space: Foundation, Evolution, Mechanism, Ability, and Outlook https://huggingface.co/papers/2604.02029 3. Generative World Renderer https://huggingface.co/papers/2604.02329 4. SKILL0: In-Context Agentic Reinforcement Learning for Skill In...

【本日の論文】 1. DataFlex: A Unified Framework for Data-Centric Dynamic Training of Large Language Models https://huggingface.co/papers/2603.26164 2. The Latent Space: Foundation, Evolution, Mechanism, Ability, and Outlook https://huggingface.co/papers/2604.02029 3. Generative World Renderer https://huggingface.co/papers/2604.02329 4. SKILL0: In-Context Agentic Reinforcement Learning for Skill In...

【本日の論文】 1. DataFlex: A Unified Framework for Data-Centric Dynamic Training of Large Language Models https://huggingface.co/papers/2603.26164 2. The Latent Space: Foundation, Evolution, Mechanism, Ability, and Outlook https://huggingface.co/papers/2604.02029 3. Generative World Renderer https://huggingface.co/papers/2604.02329 4. SKILL0: In-Context Agentic Reinforcement Learning for Skill In...

【本日の論文】 1. ClawKeeper: Comprehensive Safety Protection for OpenClaw Agents Through Skills, Plugins, and Watchers https://huggingface.co/papers/2603.24414 2. Terminal Agents Suffice for Enterprise Automation https://huggingface.co/papers/2604.00073 3. MiroEval: Benchmarking Multimodal Deep Research Agents in Process and Outcome https://huggingface.co/papers/2603.28407 4. ViGoR-Bench: How Far...

【本日の論文】 1. FIPO: Eliciting Deep Reasoning with Future-KL Influenced Policy Optimization https://huggingface.co/papers/2603.19835 2. CARLA-Air: Fly Drones Inside a CARLA World -- A Unified Infrastructure for Air-Ground Embodied Intelligence https://huggingface.co/papers/2603.28032 3. LongCat-Next: Lexicalizing Modalities as Discrete Tokens https://huggingface.co/papers/2603.27538 4. Lingshu-...

【本日の論文】 1. TAPS: Task Aware Proposal Distributions for Speculative Sampling https://huggingface.co/papers/2603.27027 2. Towards a Medical AI Scientist https://huggingface.co/papers/2603.28589 3. Gen-Searcher: Reinforcing Agentic Search for Image Generation https://huggingface.co/papers/2603.28767 4. Emergent Social Intelligence Risks in Generative Multi-Agent Systems https://huggingface.co/...

【本日の論文】 1. Out of Sight but Not Out of Mind: Hybrid Memory for Dynamic Video World Models https://huggingface.co/papers/2603.25716 2. ShotStream: Streaming Multi-Shot Video Generation for Interactive Storytelling https://huggingface.co/papers/2603.25746 3. PackForcing: Short Video Training Suffices for Long Video Sampling and Long Context Inference https://huggingface.co/papers/2603.25730 4...

【本日の論文】 1. PixelSmile: Toward Fine-Grained Facial Expression Editing https://huggingface.co/papers/2603.25728 2. Intern-S1-Pro: Scientific Multimodal Foundation Model at Trillion Scale https://huggingface.co/papers/2603.25040 3. Calibri: Enhancing Diffusion Transformers via Parameter-Efficient Calibration https://huggingface.co/papers/2603.24800 4. RealRestorer: Towards Generalizable Real-W...

【本日の論文】 1. PixelSmile: Toward Fine-Grained Facial Expression Editing https://huggingface.co/papers/2603.25728 2. Intern-S1-Pro: Scientific Multimodal Foundation Model at Trillion Scale https://huggingface.co/papers/2603.25040 3. Calibri: Enhancing Diffusion Transformers via Parameter-Efficient Calibration https://huggingface.co/papers/2603.24800 4. RealRestorer: Towards Generalizable Real-W...

【本日の論文】 1. PixelSmile: Toward Fine-Grained Facial Expression Editing https://huggingface.co/papers/2603.25728 2. Intern-S1-Pro: Scientific Multimodal Foundation Model at Trillion Scale https://huggingface.co/papers/2603.25040 3. RealRestorer: Towards Generalizable Real-World Image Restoration with Large-Scale Image Editing Models https://huggingface.co/papers/2603.25502 4. Calibri: Enhancin...

【本日の論文】 1. CUA-Suite: Massive Human-annotated Video Demonstrations for Computer-Use Agents https://huggingface.co/papers/2603.24440 2. EVA: Efficient Reinforcement Learning for End-to-End Video Agent https://huggingface.co/papers/2603.22918 3. UI-Voyager: A Self-Evolving GUI Agent Learning via Failed Experience https://huggingface.co/papers/2603.24533 4. T-MAP: Red-Teaming LLM Agents with T...

【本日の論文】 1. MinerU-Diffusion: Rethinking Document OCR as Inverse Rendering via Diffusion Decoding https://huggingface.co/papers/2603.22458 2. WildWorld: A Large-Scale Dataset for Dynamic World Modeling with Actions and Explicit State toward Generative ARPG https://huggingface.co/papers/2603.23497 3. SpecEyes: Accelerating Agentic Multimodal LLMs via Speculative Perception and Planning https:...

【本日の論文】 1. Omni-WorldBench: Towards a Comprehensive Interaction-Centric Evaluation for World Models https://huggingface.co/papers/2603.22212 2. Speed by Simplicity: A Single-Stream Architecture for Fast Audio-Video Generative Foundation Model https://huggingface.co/papers/2603.21986 3. LongCat-Flash-Prover: Advancing Native Formal Reasoning via Agentic Tool-Integrated Reinforcement Learning...

【本日の論文】 1. HopChain: Multi-Hop Data Synthesis for Generalizable Vision-Language Reasoning https://huggingface.co/papers/2603.17024 2. Astrolabe: Steering Forward-Process Reinforcement Learning for Distilled Autoregressive Video Models https://huggingface.co/papers/2603.17051 3. TerraScope: Pixel-Grounded Visual Reasoning for Earth Observation https://huggingface.co/papers/2603.19039 4. Proa...

【本日の論文】 1. Generation Models Know Space: Unleashing Implicit 3D Priors for Scene Understanding https://huggingface.co/papers/2603.19235 2. SAMA: Factorized Semantic Anchoring and Motion Alignment for Instruction-Guided Video Editing https://huggingface.co/papers/2603.19228 3. 3DreamBooth: High-Fidelity 3D Subject-Driven Video Generation Model https://huggingface.co/papers/2603.18524 4. FAST...

【本日の論文】 1. Generation Models Know Space: Unleashing Implicit 3D Priors for Scene Understanding https://huggingface.co/papers/2603.19235 2. SAMA: Factorized Semantic Anchoring and Motion Alignment for Instruction-Guided Video Editing https://huggingface.co/papers/2603.19228 3. 3DreamBooth: High-Fidelity 3D Subject-Driven Video Generation Model https://huggingface.co/papers/2603.18524 4. FAST...

【本日の論文】 1. Generation Models Know Space: Unleashing Implicit 3D Priors for Scene Understanding https://huggingface.co/papers/2603.19235 2. SAMA: Factorized Semantic Anchoring and Motion Alignment for Instruction-Guided Video Editing https://huggingface.co/papers/2603.19228 3. FASTER: Rethinking Real-Time Flow VLAs https://huggingface.co/papers/2603.19199 4. 3DreamBooth: High-Fidelity 3D Sub...