arxiv:2411.07618
hanqi yan
hanqiyan
AI & ML interests
None yet
Recent Activity
upvoted
a
paper
24 days ago
When Thinking Backfires: Mechanistic Insights Into Reasoning-Induced
Misalignment
upvoted
a
paper
about 2 months ago
Analysing Chain of Thought Dynamics: Active Guidance or Unfaithful
Post-hoc Rationalisation?
upvoted
a
paper
2 months ago
CODI: Compressing Chain-of-Thought into Continuous Space via
Self-Distillation
Organizations
None yet