HuggingFace
ScholarCatalyst: A Benchmark for Retrieving Papers That Inspire New Research
What makes great scientists great? Even as AI systems start to make progress on open problems, scientists remain far ...
🤖 AI 前沿资讯
每天自动聚合 AI 领域最新动态
HuggingFace
Keyword Harnesses Fail Open: A Cheap Diagnostic Ladder for Tool-Use Claims in Small Language Models
Keyword-matching benchmarks can credit small models for tool use they never perform. We document such a false positiv...
HuggingFace
OpenTumorBoard: A Real-World Benchmark of Multidisciplinary Tumor Board Discussion Trajectories
Multidisciplinary tumor boards integrate multimodal clinical observations and longitudinal patient histories through ...
HuggingFace
X-Tree: Tokenizing Reusable Experience for Efficient Agent Generalization
Multi-step agents are trained on flat action streams: SFT and RLVR weight every token uniformly and ignore the sub-pr...
HuggingFace
Honeycomb: Constant-Size Scene Memory Representation for Video World Models
Video world models require persistent scene memory to maintain consistency during long-horizon video generation. Exis...
HuggingFace
Transformers Stop Thinking Too Early, and a Tiny LoRA Fixes It
Pretrained transformers use little of their depth to follow references in context. Thirteen base models reliably foll...
HuggingFace
Persona Dosing: Calibrated Activation Steering for Graded Trait Control
An activation-steering coefficient sets intervention strength, but requesting a particular degree of persona expressi...
HuggingFace
Video Generation Models: A Survey of Post-Training and Alignment
Video generation has rapidly progressed from short, low-quality clips to high-resolution, long-duration sequences wit...
HuggingFace
MemFold: Learning Compact Soft Memory for Long-Context Personalization via On-Policy Optimization
An assistant that serves the same user over a long horizon has to answer from what that user has revealed: which pref...
HuggingFace
Ego2Act: Evaluating Goal-Directed Manipulation in Egocentric Video Generation
Video generation models are increasingly being explored as world simulators for embodied planning and learning. To do...
arXiv
From Knowledge Access to Source Learning: Developing Source-Specific Competence
Large language model (LLM) agents increasingly rely on persistent external sources to solve sequences of knowledge-in...
arXiv
DuoMind: Enabling Distributed Multi-Robot Coordination with Semantic Communication
Vision-language models (VLMs) and vision-language-action models (VLAs) have recently driven rapid progress in general...