AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

Post Hoc Explanations of Language Models Can Improve Language Models

Satyapriya Krishna, Jiaqi Ma, Dylan Z Slack, Asma Ghandeharioun, Sameer Singh, Himabindu Lakkaraju

NeurIPS 2023 · Large language models › Language, safety and interpretability · approach: Multilingual LLM data and culture · 18 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace human CoT rationales with post hoc explanation keyword rationalesGPT-3.5 · 7-task averageaccuracy: 62.9 -> 72.7Improved
replace human CoT rationales with post hoc explanation rationalesGPT-3 · 7-task averageaccuracy: 58.0 -> 67.2Improved
replace CoT with AMPLIFY on a math multi-step taskGPT-3.5 · GSM8kaccuracy: 43.5 -> 27.4Hurt
replace CoT with AMPLIFY on word sortingGPT-3.5 · Word Sortingaccuracy: 43.1 -> 43.6No clear effect
fine-tune GPT-2 proxy for 200 epochs instead of none (ablation)AMPLIFY · 7-task average (GPT-3.5)accuracy: 65.4 -> 72.7Improved
use highest-MCS samples instead of random samples (ablation)AMPLIFY · 7-task average (GPT-3.5)accuracy: 62.0 -> 72.7Improved
use highest-MCS instead of lowest-MCS samples (ablation)AMPLIFY · 7-task average (GPT-3.5)accuracy: 69.8 -> 72.7Improved
use highest-MCS instead of faithful-explanation sample selection (ablation)AMPLIFY · 7-task average (GPT-3.5)accuracy: 69.7 -> 72.7Improved
Gradient x Input instead of vanilla gradient explanations (ablation)AMPLIFY · 7-task average (GPT-3.5)accuracy: 68.6 -> 72.5Improved
use contrastive variants of gradient explanations (ablation)AMPLIFY · 7-task average (GPT-3.5)accuracy: 70.1 vs 68.6/72.5No clear effect
increase keywords k and shots s from (2,5) to (5,10) (ablation)AMPLIFY · 7-task average (GPT-3.5)accuracy: 66.0 -> 72.6Improved
tailor rationale template to the question on Causal Judgment (ablation)AMPLIFY · Causal Judgmentaccuracy: 60.5 -> 63.1Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy