Daily Papers of 2023-07-21

1. TokenFlow: Consistent Diffusion Features for Consistent Video Editing 58 upvotes, #1 of 2023-07-21
2. Meta-Transformer: A Unified Framework for Multimodal Learning 45 upvotes, #2 of 2023-07-21
3. Brain2Music: Reconstructing Music from Human Brain Activity 42 upvotes, #3 of 2023-07-21
4. FLASK: Fine-grained Language Model Evaluation based on Alignment Skill Sets 14 upvotes, #4 of 2023-07-21
5. Improving Multimodal Datasets with Image Captioning 12 upvotes, #5 of 2023-07-21
6. PASTA: Pretrained Action-State Transformer Agents 11 upvotes, #6 of 2023-07-21
7. SciBench: Evaluating College-Level Scientific Problem-Solving Abilities of Large Language Models 9 upvotes, #7 of 2023-07-21
8. The Role of Entropy and Reconstruction in Multi-View Self-Supervised Learning 8 upvotes, #8 of 2023-07-21
9. Instruction-following Evaluation through Verbalizer Manipulation 4 upvotes, #9 of 2023-07-21

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.