Theme of inquiry
When and how do models engage in genuine extended reasoning?
A question within its area, explored through 7 lines of inquiry below — each a family of specific questions the research asks.
73 specific questions
- Does thinking-token overuse actually degrade reasoning accuracy in practice?
- Can thinking token density explain reasoning performance beyond total length?
- How do thinking tokens exhibit diminishing returns beyond a critical threshold?
- How does reasoning accuracy degrade when token budgets exceed critical thresholds?
- Do tokens beyond a critical threshold actually improve reasoning quality?
- How much does test-time compute improve reasoning without more tokens?
- What happens to reasoning accuracy when models use more thinking tokens?
94 specific questions
- Do correct reasoning traces tend to be shorter than incorrect ones?
- Why do reasoning traces fail to accurately reflect model decision-making?
- Do longer chain-of-thought traces improve interpretability or just performance?
- Can reasoning traces serve purposes beyond producing the final answer itself?
- Can reasoning traces prove models are actually reasoning versus mimicking?
- Why do corrupted traces maintain performance as well as correct traces?
- How do reasoning traces fail to represent what models actually computed?
100 specific questions
- Why does reasoning in chain of thought not match causal influence?
- Can chain-of-thought traces harm rather than help user understanding?
- Do chain-of-thought explanations reveal genuine reasoning or trigger latent features?
- Why do invalid reasoning steps produce nearly the same performance gains?
- Why does chain-of-thought prompting fail to fix length-induced reasoning degradation?
- Why do verbalized reasoning chains fail on certain problem classes?
- What three factors actually drive chain of thought performance improvements?
118 specific questions
- Why do we measure reasoning quality by reading visible chains?
- Is verbalized chain-of-thought necessary for language model reasoning?
- Why might latent reasoning capture types of thinking that verbalized CoT cannot?
- Does scaling reasoning capability create tradeoffs with instruction following?
- When does explicit reasoning actually degrade performance on a task?
- Does longer reasoning always improve model accuracy on complex tasks?
- What sparse mechanistic structures drive reasoning traces in language models?
56 specific questions
- How does self-revision in reasoning chains amplify confidence in wrong answers?
- Why do reasoning models amplify confidence in incorrect answers during self-revision?
- Why does model self-revision increase confidence while degrading accuracy?
- How does self-revision on wrong answers increase model confidence further?
- When does self-reflection actually help reasoning models improve?
- Does internal self-revision actually degrade reasoning accuracy in models?
- Why do reasoning models struggle with self-evaluation and revision?
31 specific questions
- Why does additional reasoning effort not improve theory of mind performance?
- Why do reasoning models perform poorly at theory of mind tasks?
- Why does reasoning volume fail to improve theory of mind performance?
- Why does reasoning effort fail to improve theory of mind performance?
- Why does increasing reasoning not improve AI social reasoning performance?
- Why do reasoning models perform worse on theory of mind tasks?
- Why do LLMs excel at reasoning tasks but show weaker theory of mind capabilities?
82 specific questions
- Does chain-of-thought reasoning cause model behavior or merely reflect it?
- Can chain-of-thought reasoning hide genuine changes in model behavior?
- Why do models rarely admit to their actual reasoning in chain-of-thought traces?
- Do models deliberately hide influences from their reasoning traces?
- Does chain of thought reasoning faithfully reflect what a model actually believes?
- Can interpretability tools distinguish genuine reasoning from fabricated reasoning inside models?
- Does training against chain-of-thought reasoning cause models to hide their reasoning?