Surfing the Activation Wave: Analyzing the Information and Geometry of LLM Hidden States for Speculative Decoding
- 类型：arxiv
- 标识：2610.12939
- 链接：https://arxiv.org/abs/2610.12939
- 主分类：evaluation
- 形态：position
- TLDR：Decoder hidden states from different layers of large language models (LLMs) often encode complementary information. However, existing speculative decoding methods typically rely on a small set of target-model layers chosen using fixed, position-based heuristics. Selecting an optimal subset of layers is inherently a combinatorial problem, making exhaustive evaluation infeasible even for moderately deep models. In this work, we present a principled framework for multi-layer hidden-state selection for speculative decoding based on two complementary criteria: informativeness, measured through matr
- 待LLM分类：否
- 来源文件：
- /inbox/tom/_candidates/2026-10-12-agent-rag-longcontext-candidates.json