Daily Papers of 2024-05-30

  1. MAP-Neo: Highly Capable and Transparent Bilingual Large Language Model Series 41 upvotes, #1 of 2024-05-30
  2. T2V-Turbo: Breaking the Quality Bottleneck of Video Consistency Model with Mixed Reward Feedback 18 upvotes, #2 of 2024-05-30
  3. LLMs achieve adult human performance on higher-order theory of mind tasks 14 upvotes, #3 of 2024-05-30
  4. Nearest Neighbor Speculative Decoding for LLM Generation and Attribution 12 upvotes, #4 of 2024-05-30
  5. Self-Exploring Language Models: Active Preference Elicitation for Online Alignment 12 upvotes, #4 of 2024-05-30
  6. EasyAnimate: A High-Performance Long Video Generation Method based on Transformer Architecture 11 upvotes, #6 of 2024-05-30
  7. Offline Regularised Reinforcement Learning for Large Language Models Alignment 11 upvotes, #6 of 2024-05-30
  8. Zipper: A Multi-Tower Decoder Architecture for Fusing Modalities 10 upvotes, #8 of 2024-05-30
  9. NPGA: Neural Parametric Gaussian Avatars 9 upvotes, #9 of 2024-05-30
  10. Value-Incentivized Preference Optimization: A Unified Approach to Online and Offline RLHF 8 upvotes, #10 of 2024-05-30
  11. SoundCTM: Uniting Score-based and Consistency Models for Text-to-Sound Generation 7 upvotes, #11 of 2024-05-30
  12. Atlas3D: Physically Constrained Self-Supporting Text-to-3D for Simulation and Fabrication 5 upvotes, #12 of 2024-05-30

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.