Daily Papers of 2023-11-20

1. Video-LLaVA: Learning United Visual Representation by Alignment Before Projection 28 upvotes, #1 of 2023-11-20
2. Rethinking Attention: Exploring Shallow Feed-Forward Neural Networks as an Alternative to Attention Layers in Transformers 25 upvotes, #2 of 2023-11-20
3. Emu Video: Factorizing Text-to-Video Generation by Explicit Image Conditioning 25 upvotes, #2 of 2023-11-20
4. Camels in a Changing Climate: Enhancing LM Adaptation with Tulu 2 19 upvotes, #4 of 2023-11-20
5. MetaDreamer: Efficient Text-to-3D Creation With Disentangling Geometry and Texture 17 upvotes, #5 of 2023-11-20
6. SelfEval: Leveraging the discriminative nature of generative models for evaluation 17 upvotes, #5 of 2023-11-20
7. Testing Language Model Agents Safely in the Wild 11 upvotes, #7 of 2023-11-20
8. I&S-ViT: An Inclusive & Stable Method for Pushing the Limit of Post-Training ViTs Quantization 9 upvotes, #8 of 2023-11-20
9. VideoCon: Robust Video-Language Alignment via Contrast Captions 8 upvotes, #9 of 2023-11-20
10. Distilling and Retrieving Generalizable Knowledge for Robot Manipulation via Language Corrections 7 upvotes, #10 of 2023-11-20
11. UnifiedVisionGPT: Streamlining Vision-Oriented AI through Generalized Multimodal Framework 5 upvotes, #11 of 2023-11-20

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.