Artisan在逾10个代表模型中碰撞率最低

重点关注

• RSI-Anything: Unlocking Recursive Self-Improvement with World Models score 8
  入选 HF Daily Papers；HF 热度: 11 upvotes (+3)；有代码实现
• Towards Physically Grounded Articulated 3D Assets: A Multi-Agent Pipeline and Benchmark score 8
  入选 HF Daily Papers；HF 热度: 5 upvotes (+2)；有代码实现；关键词(1): agentic
• ProgramGym: Scaling Training Environments for Long-Horizon Program Reproduction Agents score 7
  入选 HF Daily Papers；HF 热度: 15 upvotes (+3)；关键词(3): scaling, fine-tuning, coding
• EmbodiedCLBench: Evaluating Continual Learning for Self-Evolving Embodied Agents score 7
  入选 HF Daily Papers；HF 热度: 2 upvotes (+1)；有代码实现；关键词(1): embodied
• Learning Representations to Guide Pixel Diffusion Transformers score 6
  入选 HF Daily Papers；有代码实现；关键词(2): scaling, text-to-image
• Rethinking Relative Policy Differences Transfer of On-Policy Distillation score 6
  入选 HF Daily Papers；有代码实现；关键词(2): distillation, fine-tuning
• AMLworld-Compact: Importance-Weighted Downsampling for Cost-Effective LLM Evaluation and Error Diagnosis score 5
  入选 HF Daily Papers；有代码实现

也值得关注

• When Charts Deceive Web Agents: A Paired Benchmark for Misleading Visualization-Grounded Tasks score 4
  关键词(1): vision-language；顶会接收: EMNLP
• OGPO: Offline Goal-conditioned Policy Optimization for Recoverable Vision-Language-Action Models score 4
  关键词(3): fine-tuning, post-training, vision-language；顶会接收: NeurIPS
• Masked Diffusion Language Agents for Tool-Integrated Chemical Reasoning score 4
  关键词(4): fine-tuning, post-training, agentic, reasoning；顶会接收: NeurIPS
• ELMA: Benchmarking Anaphoric Compositions for Long-Term Text-to-Motion Generation score 4
  关键词(1): fine-tune；顶会接收: NeurIPS
• zip2zip++: Towards Inference-Time Token Compression with Minimal Quality Loss score 4
  机构: EPFL；关键词(2): compression, reasoning
• X-AVDD: Cross-Attentive Audio-Visual Dataset Distillation score 4
  关键词(2): distillation, synthetic data；顶会接收: NeurIPS
• When Edge Independence Fails: Joint Graph Diffusion with Latent Sociability Priors score 4
  关键词(1): edge；顶会接收: NeurIPS
• Your Discrete Graph Diffusion Model is Secretly Memorizing score 4
  关键词(1): edge；顶会接收: NeurIPS
• Where to Look Matters: Rethinking Sub-Volume Sampling in 3D Medical Self-supervised Learning score 4
  关键词(2): lightweight, pretraining；顶会接收: NeurIPS
• AI Models Can Provably Hide Arbitrary Capabilities score 4
  关键词(2): deployment, fine-tuning；顶会接收: NeurIPS
• Why Heavy-Tailed Weight Spectra Predict Model Quality score 3
  顶会接收: NeurIPS
• Opening the Black Box of Classifier-Free Guidance via Information Bottleneck score 3
  顶会接收: NeurIPS
• Towards the Roof of the World: NLP Progress in Tibetan Languages score 3
  顶会接收: ACL
• PANDA: Prior-guided Attentional Dual-path Architecture score 3
  顶会接收: NeurIPS