Song

Song on Hugging Face Daily Papers: 25 papers, 9 in the top 3 of their day, 1,414 upvotes.

  1. SoL-Refiner: Speed-of-Light One-Step Refinement for High-Resolution Video 41 upvotes, #27 of 2026-09-30
  2. SoL-Pi: Recursively Scaling Auto-Research Loops for Efficient Agent Harness 130 upvotes, #2 of 2026-09-18
  3. Sol-Attn: Accelerating Video Generation Inference via On-the-Fly Attention Sparsification 35 upvotes, #8 of 2026-07-28
  4. SANA-Video 2.0: Hybrid Linear Attention with Attention Residuals for Efficient Video Generation 39 upvotes, #5 of 2026-07-24
  5. Nemotron-Labs-Diffusion: A Tri-Mode Language Model Unifying Autoregressive, Diffusion, and Self-Speculation Decoding 12 upvotes, #18 of 2026-07-08
  6. Cosmos 3: Omnimodal World Models for Physical AI 115 upvotes, #1 of 2026-06-04
  7. LongLive-RAG: A General Retrieval-Augmented Framework for Long Video Generation 19 upvotes, #17 of 2026-06-02
  8. SANA-Streaming: Real-time Streaming Video Editing with Hybrid Diffusion Transformer 36 upvotes, #12 of 2026-06-01
  9. LongLive-2.0: An NVFP4 Parallel Infrastructure for Long Video Generation 109 upvotes, #3 of 2026-05-19
  10. SANA-WM: Efficient Minute-Scale World Modeling with Hybrid Linear Diffusion Transformer 80 upvotes, #4 of 2026-05-15
  11. AnyFlow: Any-Step Video Diffusion Model with On-Policy Flow Map Distillation 96 upvotes, #3 of 2026-05-14
  12. Quant VideoGen: Auto-Regressive Long Video Generation via 2-Bit KV-Cache Quantization 33 upvotes, #8 of 2026-02-05
  13. DC-VideoGen: Efficient Video Generation with Deep Compression Video Autoencoder 33 upvotes, #8 of 2025-10-01
  14. SANA-Video: Efficient Video Generation with Block Linear Diffusion Transformer 38 upvotes, #7 of 2025-09-30
  15. LongLive: Real-time Interactive Long Video Generation 170 upvotes, #1 of 2025-09-29
  16. XAttention: Block Sparse Attention with Antidiagonal Scoring 12 upvotes, #23 of 2025-03-21
  17. SANA-Sprint: One-Step Diffusion with Continuous-Time Consistency Distillation 24 upvotes, #12 of 2025-03-14
  18. LServe: Efficient Long-sequence LLM Serving with Unified Sparse Attention 12 upvotes, #15 of 2025-02-21
  19. SANA 1.5: Efficient Scaling of Training-Time and Inference-Time Compute in Linear Diffusion Transformer 16 upvotes, #9 of 2025-01-31
  20. Wolf: Captioning Everything with a World Summarization Framework 30 upvotes, #2 of 2024-07-29
  21. VILA^2: VILA Augmented VILA 36 upvotes, #2 of 2024-07-25
  22. BitDelta: Your Fine-Tune May Only Be Worth One Bit 20 upvotes, #8 of 2024-02-16
  23. VILA: On Pre-training for Visual Language Models 21 upvotes, #3 of 2023-12-13
  24. PockEngine: Sparse and Efficient Fine-tuning in a Pocket 15 upvotes, #4 of 2023-10-30
  25. LongLoRA: Efficient Fine-tuning of Long-Context Large Language Models 90 upvotes, #1 of 2023-09-22

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.