UniCog: Uncovering Cognitive Abilities of LLMs through Latent Mind Space Analysis
This work addresses the interpretability gap in LLM cognition for researchers and developers, offering a new paradigm but is incremental in building on existing latent variable models.
The paper tackled the problem of understanding how cognitive abilities are engaged during LLM reasoning by proposing UniCog, a framework that analyzes LLM cognition through a latent mind space, revealing a Pareto principle and improving reasoning performance by up to 7.5% on benchmarks.
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.