Daily Papers of 2024-04-04

1. Mixture-of-Depths: Dynamically allocating compute in transformer-based language models 88 upvotes, #1 of 2024-04-04
2. Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction 54 upvotes, #2 of 2024-04-04
3. Language Models as Compilers: Simulating Pseudocode Execution Improves Algorithmic Reasoning in Language Models 44 upvotes, #3 of 2024-04-04
4. InstantStyle: Free Lunch towards Style-Preserving in Text-to-Image Generation 18 upvotes, #4 of 2024-04-04
5. ChatGLM-Math: Improving Math Problem-Solving in Large Language Models with a Self-Critique Pipeline 18 upvotes, #4 of 2024-04-04
6. On the Scalability of Diffusion-based Text-to-Image Generation 16 upvotes, #6 of 2024-04-04
7. Cross-Attention Makes Inference Cumbersome in Text-to-Image Diffusion Models 11 upvotes, #7 of 2024-04-04
8. Freditor: High-Fidelity and Transferable NeRF Editing by Frequency Decomposition 9 upvotes, #8 of 2024-04-04

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.