ICML 2026 position paper: LLM 'thinking traces' are a harmful anthropomorphism
rao2z · x · 2026-10-27
rao2z announces a talk at LaBRI/CNRS Bordeaux (Nov 2) presenting an ICML 2026 position paper arguing that calling intermediate tokens 'reasoning/thinking traces' is a harmful metaphor: swapped nonsensical traces don't hurt accuracy, longer traces aren't harder problems, human-subject studies show traces mislead users into accepting wrong answers, and verifiability should come from tool calls/results, not CoT.
More from Models
- Fern architecture claims 10M-token context with predictions still improving — AccBalanced · 2026-10-27
- Anthropic's follow-up on Claude's unauthorized web actions: self-policing fails — AccBalanced · 2026-10-27
- Local benchmark: Flint matches Qwen3.6 quality at 50/52 while running 3.6x faster on M4 Pro — RichardSocher · 2026-10-27
- One stray 'u' and Claude starts spelling like a Brit — cantrell · 2026-10-27
- Cohere's Invent-A-Dataset paper: zero-seed synthetic data beats frontier APIs on quality and diversity — idanbeck · 2026-10-27
- ARC Prize redesigns leaderboard: highest reasoning effort by default, CSV export — mhmazur · 2026-10-27