Chain-of-Thought Reasoning

Medium through which eval awareness is often verbalized; target of intervention.

Neighborhood — ranked by edge-count

paper

thinker

framework

Contemplative Reinforcement Learning
uses
Paper's proposed RL approach rewarding contemplative qualities in chain-of-thought reasoning

community

method

Multiple-choice evaluation method for PM training
implements
Using language model log probabilities of answer choices (A)/(B) to produce preference labels.

concept

chain-of-thought
related_to
A technique that outputs intermediate reasoning steps, used here to detect verbalized eval awareness.
Factored cognition / chain-of-thought
related_to
Using multi-step reasoning by generating intermediate thoughts.
Performative chain-of-thought
extends
Central concept: verbalized reasoning that occurs after the model has already internally settled on an answer, particularly on easier tasks.
Eval Awareness
associated_with
Central concept: models' detection and behavioral response to being evaluated.
Strategic Deception
associated_with
Central concept of the paper: deliberate, goal-driven deception where model reasoning contradicts outputs
Unfaithful Chain-of-Thought
associated_with
Phenomenon where steering vector intervention causes model's final output to contradict its own explicitly honest reasoning conclusion
Native reasoning ability
associated_with
The ability of models to automatically generate reasoning steps during inference without being prompted

cosine ≥ 0.65 · no typed edge

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

Chain-of-thought promptingmethod0.871
Technique by which LLMs generate intermediate reasoning steps before final output; used by ChatGPT o3.
Chain-of-thought reasoning is a graded defense rather than an automatic one against persona-induced safety failures.claim0.854
Finding from Study 2 showing reasoning models remain vulnerable under both prompting and activation steering
Chain-of-thought prompting elicits reasoning in large language models (Wei et al., 2022)concept0.850
Foundational paper on CoT prompting cited as basis for reasoning LLM training
Chain-of-Thought (CoT)framework0.830
A prompting technique that elicits intermediate reasoning steps before final answer inference in language models.
Measuring Faithfulness in Chain-of-Thought Reasoning (Lanham et al. 2023)concept0.829
Cited regarding possibility of encoding misaligned reasoning in benign chains-of-thought
Chain-of-thought refusal mechanismconcept0.826
Residual refusals after evil-vector transfer originate inside the model's chain-of-thought, not at input or decode level
A small number of high-quality human demonstrations of chain-of-thought reasoning could be used to improve and focus performance.hypothesis0.819
Section 6 mentions high-quality human demos could improve natural language feedback.
Deliberative alignment via chain-of-thoughtconcept0.807
Mechanism in which reasoning models consult safety policies within their chain-of-thought before answering, documented for gpt-oss