Cognitive Biases in Large Language Models - LessWrong
Humans, one might say, are the cyanobacteria of AI: we constantly emit large amounts of structured data, which implicitly rely on logic, causality, object permanence, history—all of that good stuff.…
x Cognitive Biases in Large Language Models — LessWrong Language Models (LLMs) AI Frontpage 18 Cognitive Biases in Large Language Models by Jan 25th Sep 2021 AI Alignment Forum Linkpost for universalprior.substack.com 15 min read 3 18 Ω 8 Humans, one might say, are the cyanobacteria of AI: we constantly emit large amounts of structured data, which implicitly rely on logic, causality, object permanence, history—all of that good stuff. All of that is implicit and encoded into our writings and videos and ‘data exhaust’. A model learning to predict must learn to understand all of that to get the b
saved by
related reading
- Do large language models understand us? | by Blaise Aguera y Arcas | Mediummedium.com
- Discovering Language Model Behaviors with Model-Written Evaluations — LessWronglesswrong.com
- Thoughts on AI in academiatheinfinitesimal.substack.com
- Cognitive Bias Detection Using Advanced Prompt Engineeringarxiv.org
- [2203.02155] Training language models to follow instructions with human feedbackarxiv.org
- the case for CoT unfaithfulness is overstated — LessWronglesswrong.com
- Large Language Models Must Be Taught to Know What They Don't Knowarxiv.org
- Against LLM Reductionism — LessWronglesswrong.com
- Language Models can Solve Computer Tasksarxiv.org
- 2301.06627arxiv.org
- Thoughts on the Alignment Implications of Scaling Language Models | Leo Gaobmk.sh
- Taking LLMs Seriously (As Language Models) — LessWronglesswrong.com