arXiv cs.CL· Jie Yu, Yuxi Sun, Zhaoxu Li, Chunxiao Liu, Xudong Jiang·· 2 天前AI 评分27
CoTS：面向大语言模型隐式推理的竞争性 token 选择
CoTS: Competitive Token Selection for Latent Reasoning in Large Language Models
AI 导读
研究者提出免训练框架 CoTS，将隐式推理干预从二元模式选择改为 token 级的连续分配，利用两个最可能候选 token 的概率竞争动态平衡离散承诺与软探索。候选明显占优时保持接近采样离散 token 以稳定推理轨迹，竞争加剧时则提升软表示贡献以保留备选路径。该方法无需人工设定切换阈值或混合系数，也不增加额外前向计算，在六个推理基准上持续优于标准 CoT 及当前最优的免训练隐式推理方法。
正文 · 原文
Abstract:Recent latent reasoning methods move beyond discrete Chain-of-Thought (CoT) by replacing sampled tokens with continuous soft embeddings or hidden states, enabling richer intermediate representations and preserving alternative reasoning paths. However, the need for latent reasoning varies across decoding steps: confident predictions favor discrete commitment, whereas ambiguous steps benefit from preserving multiple plausible alternatives. Existing approaches typically handle this variation through either global soft activation or hard switching based on uncertainty thresholds, treating latent intervention as an all-or-nothing decision rather than continuously adapting its strength. To address this limitation, we propose Competitive Token Selection (CoTS), a training-free framework that reframes latent intervention from binary mode selection into continuous allocation at the token level. CoTS uses the probability competition between the two most likely candidates to dynamically balance discrete commitment and soft exploration at each reasoning step. The second candidate represents the strongest alternative to the model's preferred continuation, so the separation between the two directly reflects the stability of the current discrete choice. When one candidate clearly dominates, CoTS keeps the representation close to the sampled discrete token to preserve a stable reasoning trajectory; as the competition strengthens, it increases the contribution of the soft representation to retain plausible alternatives. This competition-based allocation requires neither manually tuned switching thresholds nor mixing coefficients, and introduces no additional forward passes. Experiments across six reasoning benchmarks demonstrate that CoTS consistently outperforms standard CoT and state-of-the-art training-free latent reasoning methods. Our code is available at this https URL.
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.20642 [cs.CL] |
| (or arXiv:2610.20642v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.20642 arXiv-issued DOI via DataCite |
Submission history
From: Jie Yu [view email]
[v1]
Thu, 8 Oct 2026 02:47:16 UTC (117 KB)
来源：arXiv cs.CL · arxiv.org