通过潜在心智空间分析大模型认知能力，发现其推理核心与专属特征并存。
UniCog: Uncovering Cognitive Abilities of LLMs through Latent Mind Space Analysis
- 构建潜在心智空间模型，将密集激活编码为稀疏解耦的隐变量维度。
- 发现大模型推理遵循帕累托法则，共享核心与特定能力特征共存。
- 识别推理失败源于潜在激活异常，可提升性能最高7.5%。
现有研究指出大语言模型（LLMs）的认知过程与人类有本质差异，但现有可解释性方法仍难以阐明推理过程中认知能力的调用机制。本文提出UniCog，一种通过潜在心智空间分析LLM认知的统一框架。该框架将多种能力从密集模型激活中编码为稀疏、解耦的潜在维度。在包括DeepSeek-V3.2和GPT-4o在内的六种先进模型上进行广泛分析，揭示出大模型认知中的帕累托原则：存在一个共享推理核心，并辅以能力特异性签名。此外，我们发现推理失败常表现为潜在激活强度异常。这些发现开启了一种新的大模型分析范式，提供了基于认知的推理动态视角。最后，基于上述洞察，我们引入一种潜在信息引导的候选优先策略，在多个挑战性基准上使推理性能提升最高达7.5%。代码已开源。
原文摘要 · Abstract (English)
A growing body of research suggests that the cognitive processes of large language models (LLMs) differ fundamentally from those of humans. However, existing interpretability methods remain limited in explaining how cognitive abilities are engaged during LLM reasoning. In this paper, we propose UniCog, a unified framework that analyzes LLM cognition via a latent mind space. Formulated as a latent variable model, UniCog encodes diverse abilities from dense model activations into sparse, disentangled latent dimensions. Through extensive analysis on six advanced LLMs, including DeepSeek-V3.2 and GPT-4o, we reveal a Pareto principle of LLM cognition, where a shared reasoning core is complemented by ability-specific signatures. Furthermore, we discover that reasoning failures often manifest as anomalous intensity in latent activations. These findings opens a new paradigm in LLM analysis, providing a cognition grounded view of reasoning dynamics. Finally, leveraging these insights, we introduce a latent-informed candidate prioritization strategy, which improves reasoning performance by up to 7.5% across challenging benchmarks. Our code is available at https://github.com/milksalute/unicog.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。