7B Searches the Web, BVB Recovers 53.7% of Facts

Featured

• Atria Dawn: The Dawn of Agentic Superintelligence score 10
  入选 HF Daily Papers; HF 热度: 436 upvotes (+4); 有代码实现; 关键词(2): production, agentic
• ZGCM-1: A Fully Open and Extremely Efficient Foundation Model for Math and Agentic Search score 10
  入选 HF Daily Papers; HF 热度: 309 upvotes (+4); 有代码实现; 关键词(9): scaling, pruning, post-training, pre-training, agentic
• Dream-RSI: Recursive Self-Improvement through Evolving Worlds score 10
  入选 HF Daily Papers; HF 热度: 295 upvotes (+4); 有代码实现; 关键词(2): lightweight, coding
• PhysBrain 1.5: From Vision-Language Models to Physical Foundation Models score 10
  入选 HF Daily Papers; HF 热度: 177 upvotes (+4); 有代码实现; 关键词(5): fine-tuning, pre-training, vision-language, embodied, open-source
• RSIAgent: Autonomous Exploration for Recursive Self-improvement in New Environments score 10
  入选 HF Daily Papers; HF 热度: 71 upvotes (+4); 有代码实现; 关键词(1): open-source
• LynnReal-Omni: Native multi-modal Video Generation for Agentic Visual Workflows score 10
  入选 HF Daily Papers; HF 热度: 46 upvotes (+4); 有代码实现; 关键词(4): lightweight, real-time, agentic, text-to-video
• Discovery Foundation Models: Toward Open-Ended Discovery Intelligence score 10
  入选 HF Daily Papers; HF 热度: 30 upvotes (+4); 有代码实现; 关键词(3): scaling, tool use, reasoning
• BVB: Benchmarking Agentic Video Understanding via Programmatic Reconstruction in Blender score 10
  入选 HF Daily Papers; HF 热度: 24 upvotes (+4); 有代码实现; 关键词(4): lightweight, agentic, coding, reasoning
• HazardAuditor: From Executable Threats to Safer Computer-Use Agents score 9
  入选 HF Daily Papers; HF 热度: 15 upvotes (+3); 有代码实现; 关键词(1): post-training
• LLaDA-UI: Bringing Block-wise Diffusion to Vision-Language GUI Agents score 9
  入选 HF Daily Papers; HF 热度: 15 upvotes (+3); 有代码实现; 关键词(5): latency, fine-tuning, pre-training, MoE, vision-language

Also Worth Noting

• Enabling Creative Exploration for Vibe Design Agents score 5
  入选 HF Daily Papers; HF 热度: 5 upvotes (+2)
• The Router Within: Eliciting Native Skill Routing from a Frozen LLM score 4
  入选 HF Daily Papers; HF 热度: 3 upvotes (+1)
• ModaLens: Measuring Image Sensitivity in Report-Conditioned Medical VLMs score 5
  入选 HF Daily Papers; HF 热度: 2 upvotes (+1); 关键词(1): vision-language
• Lightning Weave: Improving the Accuracy-Efficiency Frontier of Reasoning Models through Capability Composition score 5
  入选 HF Daily Papers; HF 热度: 2 upvotes (+1); 关键词(4): distillation, serving, post-training, reasoning
• Learning from Reliable Negatives: Confidence-Anchored Test-Time Adaptation for GUI Grounding score 4
  机构: Zhejiang University; 关键词(1): fine-tuning
• Concept-Grounded Reasoning with Prompt-Driven Localization for Interpretable Structured Report Generation score 4
  机构: Cambridge; 关键词(1): reasoning
• V-ICAL Bench: Evaluating Video In-Context Learning for Multimodal Agents in Interactive Environments score 4
  机构: Peking University; 关键词(1): agentic
• Enabling Streaming User Transcription in Full-Duplex Speech-to-Speech Models score 4
  机构: NVIDIA; 关键词(4): lightweight, real-time, leaderboard, open-source
• Navigating Sparse Evidence: Agentic Visual RAG via Explicit Context Selection and Consolidation score 4
  机构: Baidu; 关键词(5): distillation, retrieval-augmented, RAG, agentic, reasoning
• Route Me If You Can: A Benchmark for Query Reformulation Selection score 3
  机构: University of Toronto
• Artificial entrepreneurial cognition: Locating and causally steering an opportunity recognition dial inside large language models (LLMs) score 3
  机构: Stanford
• MoveBench: A Benchmark for Global-Scale Wildlife Movement Forecasting score 3
  机构: MIT