Daily Papers of 2023-05-15

1. MEGABYTE: Predicting Million-byte Sequences with Multiscale Transformers 10 upvotes, #1 of 2023-05-15
2. Better speech synthesis through scaling 6 upvotes, #2 of 2023-05-15
3. Universal Source Separation with Weakly Labelled Data 3 upvotes, #3 of 2023-05-15
4. MMG-Ego4D: Multi-Modal Generalization in Egocentric Action Recognition 1 upvotes, #4 of 2023-05-15
5. Surfacing Biases in Large Language Models using Contrastive Input Decoding 1 upvotes, #4 of 2023-05-15
6. Optimizing Memory Mapping Using Deep Reinforcement Learning 1 upvotes, #4 of 2023-05-15
7. ArtGPT-4: Artistic Vision-Language Understanding with Adapter-enhanced MiniGPT-4 1 upvotes, #4 of 2023-05-15
8. BlendFields: Few-Shot Example-Driven Facial Modeling 1 upvotes, #4 of 2023-05-15
9. Measuring Progress in Fine-grained Vision-and-Language Understanding 1 upvotes, #4 of 2023-05-15
10. What are the Desired Characteristics of Calibration Sets? Identifying Correlates on Long Form Scientific Summarization 1 upvotes, #4 of 2023-05-15
11. Towards best practices in AGI safety and governance: A survey of expert opinion 0 upvotes, #11 of 2023-05-15

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.