Daily Papers of 2025-06-23

  1. Drag-and-Drop LLMs: Zero-Shot Prompt-to-Weights 113 upvotes, #1 of 2025-06-23
  2. Vision-Guided Chunking Is All You Need: Enhancing RAG with Multimodal Document Understanding 82 upvotes, #2 of 2025-06-23
  3. PAROAttention: Pattern-Aware ReOrdering for Efficient Sparse and Quantized Attention in Visual Generation Models 57 upvotes, #3 of 2025-06-23
  4. Hunyuan-GameCraft: High-dynamic Interactive Game Video Generation with Hybrid History Condition 50 upvotes, #4 of 2025-06-23
  5. VIKI-R: Coordinating Embodied Multi-Agent Cooperation via Reinforcement Learning 33 upvotes, #5 of 2025-06-23
  6. Optimizing Multilingual Text-To-Speech with Accents & Emotions 22 upvotes, #6 of 2025-06-23
  7. Hunyuan3D 2.5: Towards High-Fidelity 3D Assets Generation with Ultimate Details 22 upvotes, #6 of 2025-06-23
  8. DreamCube: 3D Panorama Generation via Multi-plane Synchronization 20 upvotes, #8 of 2025-06-23
  9. Machine Mental Imagery: Empower Multimodal Reasoning with Latent Visual Tokens 19 upvotes, #9 of 2025-06-23
  10. InfiniPot-V: Memory-Constrained KV Cache Compression for Streaming Video Understanding 12 upvotes, #10 of 2025-06-23
  11. Hunyuan3D 2.1: From Images to High-Fidelity 3D Assets with Production-Ready PBR Material 11 upvotes, #11 of 2025-06-23
  12. UniFork: Exploring Modality Alignment for Unified Multimodal Understanding and Generation 9 upvotes, #12 of 2025-06-23
  13. From Intention to Execution: Probing the Generalization Boundaries of Vision-Language-Action Models 8 upvotes, #13 of 2025-06-23
  14. Reranking-based Generation for Unbiased Perspective Summarization 4 upvotes, #14 of 2025-06-23
  15. MEXA: Towards General Multimodal Reasoning with Dynamic Multi-Expert Aggregation 4 upvotes, #14 of 2025-06-23
  16. Long-term Traffic Simulation with Interleaved Autoregressive Motion and Scenario Generation 4 upvotes, #14 of 2025-06-23
  17. Watermarking Autoregressive Image Generation 3 upvotes, #17 of 2025-06-23
  18. Better Language Model Inversion by Compactly Representing Next-Token Distributions 1 upvotes, #18 of 2025-06-23

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.