1/4预算Agent反超4倍暴力采样

重点关注

• LMEB: Long-horizon Memory Embedding Benchmark score 6
  入选 HF Daily Papers；HF 热度: 13 upvotes (+3)

也值得关注

• Do You See What I Am Pointing At? Gesture-Based Egocentric Video Question Answering score 4
  关键词(1): reasoning；顶会接收: CVPR
• Spend Less, Reason Better: Budget-Aware Value Tree Search for LLM Agents score 4
  入选 HF Daily Papers；关键词(4): scaling, pruning, fine-tuning, reasoning
• From Sparse to Dense: Multi-View GRPO for Flow Models via Augmented Condition Space score 5
  入选 HF Daily Papers；HF 热度: 2 upvotes (+1)；关键词(2): GRPO, text-to-image
• AVION: Aerial Vision-Language Instruction from Offline Teacher to Prompt-Tuned Network score 4
  关键词(3): lightweight, distillation, vision-language；顶会接收: CVPR
• HIFICL: High-Fidelity In-Context Learning for Multimodal Tasks score 4
  关键词(1): fine-tuning；顶会接收: CVPR
• coDrawAgents: A Multi-Agent Dialogue Framework for Compositional Image Generation score 4
  关键词(1): text-to-image；顶会接收: CVPR
• Multimodal Protein Language Models for Enzyme Kinetic Parameters: From Substrate Recognition to Conformational Adaptation score 4
  关键词(2): fine-tuning, MoE；顶会接收: CVPR
• VIRD: View-Invariant Representation through Dual-Axis Transformation for Cross-View Pose Estimation score 4
  关键词(1): robotics；顶会接收: CVPR
• daVinci-Env: Open SWE Environment Synthesis at Scale score 4
  入选 HF Daily Papers；关键词(2): reasoning, open-source
• SAVA-X: Ego-to-Exo Imitation Error Detection via Scene-Adaptive View Alignment and Bidirectional Cross View Fusion score 3
  顶会接收: CVPR
• Spectral-Geometric Neural Fields for Pose-Free LiDAR View Synthesis score 3
  顶会接收: CVPR
• Stake the Points: Structure-Faithful Instance Unlearning score 3
  顶会接收: CVPR
• L2GTX: From Local to Global Time Series Explanations score 3
  机构: xAI