'Say more than not causal': descriptive and causal mech interp work inform each other
burny_tech · x · 2026-11-06
A follow-up post in the same thread, "Say more than 'not causal'," argues that saying a result fails to establish a causal mechanism is different from saying it doesn't advance our understanding of a model — the first can be correct without implying the second.
- Descriptive findings can motivate causal hypotheses; causal experiments can reveal unexpected internal patterns or overturn earlier interpretations
- Scientific progress need not follow a single observation-to-validation sequence
- The right question is what a paper claims to teach us about the model and whether the evidence supports that claim
More from Research
- Cryptographic Assumptions Say No Efficient Algorithm Exists — So AI Can't Find One — fortnow · 2026-11-06
- Lambda and Fei-Fei Li's lab present I3L: +68.3-point success in mobile manipulation — TheZachMueller · 2026-11-06
- Embodied AI's real question isn't System-1 vs System-2, but drawing capability boundaries — ZhidingYu · 2026-11-06
- Paper passes peer review: four-layer architecture for shared organizational memory in LLM agents — letandrewcook · 2026-11-06
- Apple's GSM-Symbolic paper: renaming numbers sinks SOTA models' math scores — imjustnewatai · 2026-11-06
- Elicit to present Cochrane-based evaluation of AI-assisted systematic reviews at ISPOR — elicitorg · 2026-11-06