Daily Papers of 2025-05-14

  1. MiniMax-Speech: Intrinsic Zero-Shot Text-to-Speech with a Learnable Speaker Encoder 114 upvotes, #1 of 2025-05-14
  2. Fast Text-to-Audio Generation with Adversarial Post-Training 18 upvotes, #2 of 2025-05-14
  3. AM-Thinking-v1: Advancing the Frontier of Reasoning at 32B Scale 14 upvotes, #3 of 2025-05-14
  4. Aya Vision: Advancing the Frontier of Multilingual Multimodality 11 upvotes, #4 of 2025-05-14
  5. Bring Reason to Vision: Understanding Perception and Reasoning through Model Merging 10 upvotes, #5 of 2025-05-14
  6. Measuring General Intelligence with Generated Games 10 upvotes, #5 of 2025-05-14
  7. A Multi-Dimensional Constraint Framework for Evaluating and Improving Instruction Following in Large Language Models 9 upvotes, #7 of 2025-05-14
  8. Advancing Arabic Reverse Dictionary Systems: A Transformer-Based Approach with Dataset Construction Guidelines 5 upvotes, #8 of 2025-05-14
  9. TRAIL: Trace Reasoning and Agentic Issue Localization 5 upvotes, #8 of 2025-05-14
  10. NavDP: Learning Sim-to-Real Navigation Diffusion Policy with Privileged Information Guidance 5 upvotes, #8 of 2025-05-14
  11. SkillFormer: Unified Multi-View Video Understanding for Proficiency Estimation 4 upvotes, #11 of 2025-05-14
  12. Memorization-Compression Cycles Improve Generalization 4 upvotes, #11 of 2025-05-14
  13. Optimizing Retrieval-Augmented Generation: Analysis of Hyperparameter Impact on Performance and Efficiency 3 upvotes, #13 of 2025-05-14
  14. ViMRHP: A Vietnamese Benchmark Dataset for Multimodal Review Helpfulness Prediction via Human-AI Collaborative Annotation 1 upvotes, #14 of 2025-05-14
  15. Tests as Prompt: A Test-Driven-Development Benchmark for LLM Code Generation 0 upvotes, #15 of 2025-05-14

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.