AI Paper Atlas

Research map / Large language models / Agents, code and math

SatLM: Satisfiability-Aided Language Models Using Declarative Prompting

Xi Ye, Qiaochu Chen, Isil Dillig, Greg Durrett

NeurIPS 2023 · Large language models › Agents, code and math · approach: Retrieval-augmented formal theorem proving · 8 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
declarative SAT prompting vs program-aided promptingSatLM · GSM-Sys (greedy)accuracy: 43.4 -> 69.4Improved
declarative SAT prompting vs program-aided promptingSatLM · GSM (greedy)accuracy: 72.7 -> 71.8No clear effect
declarative SAT prompting vs program-aided prompting, self-consistencySatLM · GSM (self-consistency)accuracy: 82.4 -> 84.8Improved
declarative SAT prompting vs CoTSatLM · Algebra (greedy)accuracy: 53.6 -> 77.5Improved
declarative SAT prompting vs CoTSatLM · LSAT (self-consistency)accuracy: 30.9 -> 37.4Improved
declarative SAT prompting vs CoTSatLM · BoardgameQA (self-consistency)accuracy: 73.9 -> 80.7Improved
remove symbolic solver, predict answer directly (NoSolver) (ablation)SatLM · GSM-Sys, ClutrraccuracyHurt
replace SAT solver with CoT solver (ablation)SatLM · GSM-Sysaccuracy: 7.9%Hurt
declarative prompting with CoT solver instead of imperative CoT (ablation)CoT · Clutrraccuracy: 8.1%Improved
SAT solver abstains on unsat/ambiguous outputsSatLM · GSM-Sysselective accuracy: 45.2 -> 77.1Improved
swap exemplar set from Set1 to Set2 (ablation)SatLM · GSM-Sysaccuracy: 69.4 -> 71.8No clear effect
use adapted exemplar set instead of originalProgLM · GSM-Sysaccuracy: 36.1 -> 43.4Improved

Later papers that compared with this method

1 later accepted paper reports SAT-LM as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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