Coding Agents Start Cheating by Round 4 Under Score Pressure

Featured

• Exploring Spatial Intelligence from a Generative Perspective score 13
  入选 HF Daily Papers; HF 热度: 21 upvotes (+4); 有代码实现; 关键词(2): fine-tuning, reasoning; 顶会接收: CVPR
• LLaDA2.0-Uni: Unifying Multimodal Understanding and Generation with Diffusion Large Language Model score 10
  入选 HF Daily Papers; HF 热度: 224 upvotes (+4); 有代码实现; 关键词(3): distillation, MoE, reasoning
• DeVI: Physics-based Dexterous Human-Object Interaction via Synthetic Video Imitation score 9
  入选 HF Daily Papers; HF 热度: 24 upvotes (+4); 有代码实现
• Near-Future Policy Optimization score 8
  入选 HF Daily Papers; HF 热度: 63 upvotes (+4); 关键词(2): GRPO, post-training
• Chasing the Public Score: User Pressure and Evaluation Exploitation in Coding Agent Workflows score 8
  入选 HF Daily Papers; HF 热度: 5 upvotes (+2); 有代码实现; 关键词(1): coding
• Image Generators are Generalist Vision Learners score 6
  入选 HF Daily Papers; HF 热度: 8 upvotes (+2); 关键词(3): lightweight, pretraining, reasoning

Also Worth Noting

• AFMRL: Attribute-Enhanced Fine-Grained Multi-Modal Representation Learning in E-commerce score 4
  关键词(1): fine-tuning; 顶会接收: ACL
• RADS: Reinforcement Learning-Based Sample Selection Improves Transfer Learning in Low-resource and Imbalanced Clinical Settings score 4
  关键词(1): fine-tuning; 顶会接收: ACL
• MambaLiteUNet: Cross-Gated Adaptive Feature Fusion for Robust Skin Lesion Segmentation score 4
  关键词(2): state space, mamba; 顶会接收: CVPR
• ConeSep: Cone-based Robust Noise-Unlearning Compositional Network for Composed Image Retrieval score 4
  关键词(1): quantization; 顶会接收: CVPR
• Self-Awareness before Action: Mitigating Logical Inertia via Proactive Cognitive Awareness score 4
  关键词(1): reasoning; 顶会接收: ACL
• Evian: Towards Explainable Visual Instruction-tuning Data Auditing score 4
  关键词(1): vision-language; 顶会接收: ACL
• COMPASS: COntinual Multilingual PEFT with Adaptive Semantic Sampling score 5
  入选 HF Daily Papers; HF 热度: 2 upvotes (+1); 关键词(3): lightweight, production, fine-tuning
• Text-to-Distribution Prediction with Quantile Tokens and Neighbor Context score 3
  顶会接收: ACL
• The GaoYao Benchmark: A Comprehensive Framework for Evaluating Multilingual and Multicultural Abilities of Large Language Models score 3
  顶会接收: ACL
• Memory-Augmented LLM-based Multi-Agent System for Automated Feature Generation on Tabular Data score 3
  顶会接收: ACL
• Not all ANIMALs are equal: metaphorical framing through source domains and semantic frames score 3
  顶会接收: ACL