arXiv AI
Do Latent Tokens Think? A Causal and Adversarial Analysis of Chain-of-Continuous-Thought
Read the original on arXiv AI →The Flow has not summarised this story yet — read it at arXiv AI.
Related stories
arXiv Machine Learning
Sep 22
Assessing Adversarial Robustness of Latent Reasoning Models
arXiv:2609.22228v1 Announce Type: cross Abstract: Large language models increasingly rely on long chain-of-thought (CoT) trajectories for complex reasoning, but autoregressive generation brings subst...
arXiv AI
Sep 1
When Chain-of-Thought Fails, the Solution Hides in the Hidden States
arXiv:2604.23351v2 Announce Type: replace-cross Abstract: Whether intermediate reasoning is computationally useful or merely explanatory depends on whether chain-of-thought (CoT) tokens contain task-...
arXiv AI
Sep 1
Triggering Chain-of-Thought via Latent Feature Interventions in Large Language Models
arXiv:2601.08058v2 Announce Type: replace-cross Abstract: Chain-of-Thought (CoT) prompting often improves the reasoning performance of large language models (LLMs), but the internal signal that trigg...
arXiv Machine Learning
Aug 5
The Illusion of Superposition? A Principled Analysis of Latent Thinking in Language Models
arXiv:2604. 06374v2 Announce Type: replace-cross Abstract: Latent reasoning via continuous chain-of-thoughts (Latent CoT) has emerged as a promising alternative to discrete CoT reasoning.
arXiv AI
Jul 14
Valid $\ne$ Necessary: Diagnosing Latent Inefficiency in Chain-of-Thought
arXiv:2607. 11266v1 Announce Type: new Abstract: Chain-of-Thought (CoT) prompting has significantly advanced the reasoning capabilities of Large Language Models (LLMs), yet it often incurs substantial computational costs due to over-reasoning: the generation of redundant, verbose, or irrelevant steps.
arXiv AI
Sep 21
When Steering Fails in Latent Reasoning: A Latent-to-Language Transition Gap
The paper investigates the effectiveness of activation steering in latent chain-of-thought (CoT) reasoning compared to explicit CoT. It finds that steering continuous latent thoughts yields weaker impacts on language generation, even when hidden representations are shifted similarly. The authors propose a latent-to-language transition gap, supported by evidence of abrupt output distribution changes at the transition boundary and weaker bidirectional control in latent CoT.