Large Language Models (LLMs) struggle with complex reasoning due to limited diversity and inefficient search. We propose Soft Reasoning, 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. The code is released at https://github.com/alickzhu/Soft-Reasoning.
@article{arxiv.2505.24688,
title = {Soft Reasoning: Navigating Solution Spaces in Large Language Models through Controlled Embedding Exploration},
author = {Qinglin Zhu and Runcong Zhao and Hanqi Yan and Yulan He and Yudong Chen and Lin Gui},
journal= {arXiv preprint arXiv:2505.24688},
year = {2025}
}