2025-10-27
Models
Papers
- [2510.02361] ChunkLLM: A Lightweight Pluggable Framework for Accelerating LLMs Inference
- [2505.22618] Fast-dLLM: Training-free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decoding
- [2509.17863] Expert-as-a-Service: Towards Efficient, Scalable, and Robust Large-scale MoE Serving
- [2510.18234] DeepSeek-OCR: Contexts Optical Compression
- [2510.15870] OmniVinci: Enhancing Architecture and Data for Omni-Modal Understanding LLM
- [2510.15857] BLIP3o-NEXT: Next Frontier of Native Image Generation
- [2510.16932] Prompt-MII: Meta-Learning Instruction Induction for LLMs
Code
- [ ]
Articles
- BERT is just a Single Text Diffusion Step | nathan.rs
- the bug that taught me more about PyTorch than years of using it | Elana Simon
- Supercharge your OCR Pipelines with Open Models
- torch.compile, the missing manual - Google Docs
Videos
- [ ]