Daily Papers of 2024-11-21

  1. SageAttention2 Technical Report: Accurate 4 Bit Attention for Plug-and-play Inference Acceleration 47 upvotes, #1 of 2024-11-21
  2. VBench++: Comprehensive and Versatile Benchmark Suite for Video Generative Models 28 upvotes, #2 of 2024-11-21
  3. VideoAutoArena: An Automated Arena for Evaluating Large Multimodal Models in Video Analysis through User Simulation 17 upvotes, #3 of 2024-11-21
  4. SAMURAI: Adapting Segment Anything Model for Zero-Shot Visual Tracking with Motion-Aware Memory 16 upvotes, #4 of 2024-11-21
  5. When Precision Meets Position: BFloat16 Breaks Down RoPE in Long-Context Training 13 upvotes, #5 of 2024-11-21
  6. Is Your LLM Secretly a World Model of the Internet? Model-Based Planning for Web Agents 11 upvotes, #6 of 2024-11-21
  7. Stylecodes: Encoding Stylistic Information For Image Generation 10 upvotes, #7 of 2024-11-21
  8. ViBe: A Text-to-Video Benchmark for Evaluating Hallucination in Large Multimodal Models 7 upvotes, #8 of 2024-11-21
  9. Loss-to-Loss Prediction: Scaling Laws for All Datasets 5 upvotes, #9 of 2024-11-21
  10. Generating Compositional Scenes via Text-to-image RGBA Instance Generation 3 upvotes, #10 of 2024-11-21
  11. ORID: Organ-Regional Information Driven Framework for Radiology Report Generation 2 upvotes, #11 of 2024-11-21

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.