Daily Papers of 2025-09-16

  1. OmniWorld: A Multi-Domain and Multi-Modal Dataset for 4D World Modeling 100 upvotes, #1 of 2025-09-16
  2. UI-S1: Advancing GUI Automation via Semi-online Reinforcement Learning 45 upvotes, #2 of 2025-09-16
  3. InternScenes: A Large-scale Simulatable Indoor Scene Dataset with Realistic Layouts 30 upvotes, #3 of 2025-09-16
  4. LongEmotion: Measuring Emotional Intelligence of Large Language Models in Long-Context Interaction 26 upvotes, #4 of 2025-09-16
  5. Lost in Embeddings: Information Loss in Vision-Language Models 25 upvotes, #5 of 2025-09-16
  6. LazyDrag: Enabling Stable Drag-Based Editing on Multi-Modal Diffusion Transformers via Explicit Correspondence 18 upvotes, #6 of 2025-09-16
  7. SearchInstruct: Enhancing Domain Adaptation via Retrieval-Based Instruction Dataset Creation 16 upvotes, #7 of 2025-09-16
  8. Locality in Image Diffusion Models Emerges from Data Statistics 12 upvotes, #8 of 2025-09-16
  9. Learning to Optimize Multi-Objective Alignment Through Dynamic Reward Weighting 12 upvotes, #8 of 2025-09-16
  10. Measuring Epistemic Humility in Multimodal Large Language Models 6 upvotes, #10 of 2025-09-16
  11. Nav-R1: Reasoning and Navigation in Embodied Scenes 6 upvotes, #10 of 2025-09-16
  12. Look Again, Think Slowly: Enhancing Visual Reflection in Vision-Language Models 5 upvotes, #12 of 2025-09-16
  13. PersonaX: Multimodal Datasets with LLM-Inferred Behavior Traits 4 upvotes, #13 of 2025-09-16
  14. FuseCodec: Semantic-Contextual Fusion and Supervision for Neural Codecs 3 upvotes, #14 of 2025-09-16
  15. CognitiveSky: Scalable Sentiment and Narrative Analysis for Decentralized Social Media 3 upvotes, #14 of 2025-09-16
  16. GAPrune: Gradient-Alignment Pruning for Domain-Aware Embeddings 1 upvotes, #16 of 2025-09-16
  17. ClaimIQ at CheckThat! 2025: Comparing Prompted and Fine-Tuned Language Models for Verifying Numerical Claims 1 upvotes, #16 of 2025-09-16
  18. EthicsMH: A Pilot Benchmark for Ethical Reasoning in Mental Health AI 1 upvotes, #16 of 2025-09-16
  19. Dr.V: A Hierarchical Perception-Temporal-Cognition Framework to Diagnose Video Hallucination by Fine-grained Spatial-Temporal Grounding 1 upvotes, #16 of 2025-09-16
  20. ToolRM: Outcome Reward Models for Tool-Calling Large Language Models 1 upvotes, #16 of 2025-09-16

Data: hysts-bot-data/daily-papers-stats and the Daily Papers API. Open data: tardellirs/paper-pulse-data. Sister project: Model Pulse, the download history of every model on the Hub.