Navigating Solution Spaces in Large Language Models Through Controlled Embedding Exploration

Abstract

Large Language Models (LLMs) struggle with reasoning due to limited diversity and inefficient search. We propose an embedding-based search framework that optimises the embedding of the first token to guide generation. It combines (1) Embedding perturbation for controlled exploration and (2) Bayesian optimisation to refine embeddings via a verifier-guided objective, balancing exploration and exploitation. This approach improves reasoning accuracy and coherence while avoiding reliance on heuristic search. Experiments demonstrate superior correctness with minimal computation, making it a scalable, model-agnostic solution.

Cite

Text

Zhu et al. "Navigating Solution Spaces in Large Language Models Through Controlled Embedding Exploration." ICLR 2025 Workshops: LLM_Reason_and_Plan, 2025.

Markdown

[Zhu et al. "Navigating Solution Spaces in Large Language Models Through Controlled Embedding Exploration." ICLR 2025 Workshops: LLM_Reason_and_Plan, 2025.](https://mlanthology.org/iclrw/2025/zhu2025iclrw-navigating/)

BibTeX

@inproceedings{zhu2025iclrw-navigating,
  title     = {{Navigating Solution Spaces in Large Language Models Through Controlled Embedding Exploration}},
  author    = {Zhu, Qinglin and Zhao, Runcong and Yan, Hanqi and He, Yulan and Chen, Yudong and Gui, Lin},
  booktitle = {ICLR 2025 Workshops: LLM_Reason_and_Plan},
  year      = {2025},
  url       = {https://mlanthology.org/iclrw/2025/zhu2025iclrw-navigating/}
}