Daily Papers of 2024-10-01

  1. Addition is All You Need for Energy-efficient Language Models 109 upvotes, #1 of 2024-10-01
  2. MM1.5: Methods, Analysis & Insights from Multimodal LLM Fine-tuning 46 upvotes, #2 of 2024-10-01
  3. Ruler: A Model-Agnostic Method to Control Generated Length for Large Language Models 26 upvotes, #3 of 2024-10-01
  4. DiaSynth -- Synthetic Dialogue Generation Framework 19 upvotes, #4 of 2024-10-01
  5. Hyper-Connections 17 upvotes, #5 of 2024-10-01
  6. Cottention: Linear Transformers With Cosine Attention 14 upvotes, #6 of 2024-10-01
  7. UniAff: A Unified Representation of Affordances for Tool Usage and Articulation with Vision-Language Models 13 upvotes, #7 of 2024-10-01
  8. Scaling Proprioceptive-Visual Learning with Heterogeneous Pre-trained Transformers 11 upvotes, #8 of 2024-10-01
  9. Image Copy Detection for Diffusion Models 10 upvotes, #9 of 2024-10-01
  10. Coffee-Gym: An Environment for Evaluating and Improving Natural Language Feedback on Erroneous Code 8 upvotes, #10 of 2024-10-01
  11. Visual Question Decomposition on Multimodal Large Language Models 7 upvotes, #11 of 2024-10-01
  12. Can Models Learn Skill Composition from Examples? 7 upvotes, #11 of 2024-10-01
  13. IDEAW: Robust Neural Audio Watermarking with Invertible Dual-Embedding 1 upvotes, #13 of 2024-10-01

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.