Daily Papers of 2024-09-05

1. Loopy: Taming Audio-Driven Portrait Avatar with Long-Term Motion Dependency 81 upvotes, #1 of 2024-09-05
2. LongLLaVA: Scaling Multi-modal LLMs to 1000 Images Efficiently via Hybrid Architecture 51 upvotes, #2 of 2024-09-05
3. LongCite: Enabling LLMs to Generate Fine-grained Citations in Long-context QA 41 upvotes, #3 of 2024-09-05
4. MMMU-Pro: A More Robust Multi-discipline Multimodal Understanding Benchmark 27 upvotes, #4 of 2024-09-05
5. Arctic-SnowCoder: Demystifying High-Quality Data in Code Pretraining 16 upvotes, #5 of 2024-09-05
6. FastVoiceGrad: One-step Diffusion-Based Voice Conversion with Adversarial Conditional Diffusion Distillation 9 upvotes, #6 of 2024-09-05
7. Affordance-based Robot Manipulation with Flow Matching 8 upvotes, #7 of 2024-09-05
8. Political DEBATE: Efficient Zero-shot and Few-shot Classifiers for Political Text 7 upvotes, #8 of 2024-09-05

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.