3 papers
cs.LG2026
When Drafts Evolve: Speculative Decoding Meets Online Learning
Yu-Yang Qian, Hao-Cong Wu, Yichao Fu +2
Speculative decoding has emerged as a widely adopted paradigm for accelerating large language model inference, where a lightweight draft model rapidly generates candidate tokens th…
cs.LG2025
Efficient Long-context Language Model Training by Core Attention Disaggregation
Yonghao Zhuang, Junda Chen, Bo Pang +6
We present core attention disaggregation (CAD), a technique that improves long-context large language model training by decoupling the core attention computation, softmax(QK^T)V, f…
cs.LG2025
Efficiently Scaling LLM Reasoning with Certaindex
Yichao Fu, Junda Chen, Siqi Zhu +8
Test-time reasoning algorithms such as chain-of-thought, self-consistency, and MCTS enhance LLM problem-solving but can wastefully generate many tokens without improving accuracy.…