Daily Papers of 2024-05-27

  1. Meteor: Mamba-based Traversal of Rationale for Large Language and Vision Models 47 upvotes, #1 of 2024-05-27
  2. ConvLLaVA: Hierarchical Backbones as Visual Encoder for Large Multimodal Models 41 upvotes, #2 of 2024-05-27
  3. Grokked Transformers are Implicit Reasoners: A Mechanistic Journey to the Edge of Generalization 30 upvotes, #3 of 2024-05-27
  4. Aya 23: Open Weight Releases to Further Multilingual Progress 21 upvotes, #4 of 2024-05-27
  5. Stacking Your Transformers: A Closer Look at Model Growth for Efficient LLM Pre-Training 20 upvotes, #5 of 2024-05-27
  6. AutoCoder: Enhancing Code Large Language Model with AIEV-Instruct 19 upvotes, #6 of 2024-05-27
  7. The Road Less Scheduled 16 upvotes, #7 of 2024-05-27
  8. CraftsMan: High-fidelity Mesh Generation with 3D Native Generation and Interactive Geometry Refiner 14 upvotes, #8 of 2024-05-27
  9. Denoising LM: Pushing the Limits of Error Correction Models for Speech Recognition 11 upvotes, #9 of 2024-05-27
  10. iVideoGPT: Interactive VideoGPTs are Scalable World Models 11 upvotes, #9 of 2024-05-27
  11. Automatic Data Curation for Self-Supervised Learning: A Clustering-Based Approach 11 upvotes, #9 of 2024-05-27
  12. Data Mixing Made Efficient: A Bivariate Scaling Law for Language Model Pretraining 10 upvotes, #12 of 2024-05-27
  13. HDR-GS: Efficient High Dynamic Range Novel View Synthesis at 1000x Speed via Gaussian Splatting 4 upvotes, #13 of 2024-05-27

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.