Daily Papers of 2023-09-06

  1. One Wide Feedforward is All You Need 34 upvotes, #1 of 2023-09-06
  2. AniPortraitGAN: Animatable 3D Portrait Generation from 2D Image Collections 23 upvotes, #2 of 2023-09-06
  3. ModelScope-Agent: Building Your Customizable Agent System with Open-source Large Language Models 22 upvotes, #3 of 2023-09-06
  4. ControlMat: A Controlled Generative Approach to Material Capture 17 upvotes, #4 of 2023-09-06
  5. Efficient RLHF: Reducing the Memory Usage of PPO 16 upvotes, #5 of 2023-09-06
  6. PromptTTS 2: Describing and Generating Voices with Text Prompt 15 upvotes, #6 of 2023-09-06
  7. Hierarchical Masked 3D Diffusion Model for Video Outpainting 13 upvotes, #7 of 2023-09-06
  8. StyleAdapter: A Single-Pass LoRA-Free Model for Stylized Image Generation 12 upvotes, #8 of 2023-09-06
  9. Doppelgangers: Learning to Disambiguate Images of Similar Structures 11 upvotes, #9 of 2023-09-06
  10. Contrastive Feature Masking Open-Vocabulary Vision Transformer 10 upvotes, #10 of 2023-09-06
  11. Gated recurrent neural networks discover attention 10 upvotes, #10 of 2023-09-06
  12. MagicProp: Diffusion-based Video Editing via Motion-aware Appearance Propagation 6 upvotes, #12 of 2023-09-06
  13. Compositional Diffusion-Based Continuous Constraint Solvers 6 upvotes, #12 of 2023-09-06
  14. Diffusion Generative Inverse Design 5 upvotes, #14 of 2023-09-06
  15. Sequential Dexterity: Chaining Dexterous Policies for Long-Horizon Manipulation 4 upvotes, #15 of 2023-09-06

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.