Papers › Measuring and Narrowing the Compositionality Gap in Language Models

Measuring and Narrowing the Compositionality Gap in Language Models

7 Oct 2022arXiv:2210.03350archive 2025-07-28

Ofir Press, Muru Zhang, Sewon Min, Ludwig Schmidt, Noah A. Smith, Mike Lewis

We investigate the ability of language models to perform compositional reasoning tasks where the overall solution depends on correctly composing the answers to sub-problems. We measure how often models can correctly answer all sub-problems but not generate the overall solution, a ratio we call the compositionality gap. We evaluate this ratio by asking multi-hop questions with answers that require composing multiple facts unlikely to have been observed together during pretraining. In the GPT-3 family of models, as model size increases we show that the single-hop question answering performance improves faster than the multi-hop performance does, therefore the compositionality gap does not decrease. This surprising result suggests that while more powerful models memorize and recall more factual knowledge, they show no corresponding improvement in their ability to perform this kind of compositional reasoning. We then demonstrate how elicitive prompting (such as chain of thought) narrows the compositionality gap by reasoning explicitly. We present a new method, self-ask, that further improves on chain of thought. In our method, the model explicitly asks itself (and answers) follow-up questions before answering the initial question. We finally show that self-ask's structured prompting lets us easily plug in a search engine to answer the follow-up questions, which additionally improves accuracy.

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

ofirpress/self-ask officialmentioned in papermentioned on GitHubMIT report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Question Answering

Datasets

Introduced by this paper, per the archive.

Bamboogle

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Question Answering Bamboogle Self-ask (GPT-3; davinci-002) + Google Search Accuracy 60.0 #4 of 9 Archive leaderboard report
Question Answering Bamboogle Self-ask (GPT-3; davinci-002) Accuracy 57.6 #5 of 9 Archive leaderboard report
Question Answering Bamboogle Chain-of-Thought (GPT-3; davinci-002) Accuracy 46.4 #6 of 9 Archive leaderboard report
Question Answering Bamboogle Direct Prompting (GPT-3; davinci-002) Accuracy 17.6 #8 of 9 Archive leaderboard report
Question Answering Bamboogle Google Search Accuracy 0 #9 of 9 Archive leaderboard report
Question Answering FEVER Self-Ask EM 64.2 #2 of 8 Archive leaderboard report
Question Answering WebQuestions Self-Ask EM 31.1 #23 of 37 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

AdamAttentionAttention DropoutBPECosine AnnealingDense ConnectionsDropoutGPT-3Layer NormalizationLinear LayerLinear Warmup With Cosine AnnealingMulti-Head AttentionResidual ConnectionSoftmaxWeight Decay

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections