what is the underlying mechanism of self-reflection in reasoning LLMs?

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• The underlying mechanism of self-reflection in reasoning LLMs is not yet well understoodquestion0.964
  Broad gap motivating the entire paper
• Reasoning LLMs trigger reflection when their internal uncertainty is highhypothesis0.853
  Core hypothesis linking internal uncertainty to self-reflection behavior, tested via probing experiments
• A linear reflection direction exists in reasoning LLMs' latent representation space that governs self-reflection behaviorclaim0.839
  Core claim of ReflCtrl that a single direction captures and controls reflection
• Reflection in LLMsconcept0.822
  The core phenomenon studied: the ability of LLMs to evaluate and revise their own reasoning.
• Self-reflection consumes 25-30% of total reasoning tokens in reasoning LLMsclaim0.820
  Empirical observation motivating the need to control reflection for inference efficiency
• When LLMs produce experience claims under self-reference, is this sophisticated simulation or genuine self-representation, and how would we tell the difference?question0.801
  The core interpretive question the paper narrows but cannot definitively answer
• When does the model initiate reflection during its reasoning process?question0.797
  First central research question motivating ReflCtrl investigation
• How does reflection influence the model's reasoning performance?question0.781
  Second central research question motivating ReflCtrl investigation