Papers › OPT-IML: Scaling Language Model Instruction Meta Learning through the Lens of Generalization

OPT-IML: Scaling Language Model Instruction Meta Learning through the Lens of Generalization

22 Dec 2022arXiv:2212.12017archive 2025-07-28

Srinivasan Iyer, Xi Victoria Lin, Ramakanth Pasunuru, Todor Mihaylov, Daniel Simig, Ping Yu, Kurt Shuster, Tianlu Wang, Qing Liu, Punit Singh Koura, Xian Li, Brian O'Horo, Gabriel Pereyra, Jeff Wang, Christopher Dewan, Asli Celikyilmaz, Luke Zettlemoyer, Ves Stoyanov

Recent work has shown that fine-tuning large pre-trained language models on a collection of tasks described via instructions, a.k.a. instruction-tuning, improves their zero and few-shot generalization to unseen tasks. However, there is a limited understanding of the performance trade-offs of different decisions made during the instruction-tuning process. These decisions include the scale and diversity of the instruction-tuning benchmark, different task sampling strategies, fine-tuning with and without demonstrations, training using specialized datasets for reasoning and dialogue, and finally, the fine-tuning objectives themselves. In this paper, we characterize the effect of instruction-tuning decisions on downstream task performance when scaling both model and benchmark sizes. To this end, we create OPT-IML Bench: a large benchmark for Instruction Meta-Learning (IML) of 2000 NLP tasks consolidated into task categories from 8 existing benchmarks, and prepare an evaluation framework to measure three types of model generalizations: to tasks from fully held-out categories, to held-out tasks from seen categories, and to held-out instances from seen tasks. Through the lens of this framework, we first present insights about instruction-tuning decisions as applied to OPT-30B and further exploit these insights to train OPT-IML 30B and 175B, which are instruction-tuned versions of OPT. OPT-IML demonstrates all three generalization abilities at both scales on four different evaluation benchmarks with diverse tasks and input formats -- PromptSource, FLAN, Super-NaturalInstructions, and UnifiedSKG. Not only does it significantly outperform OPT on all benchmarks but is also highly competitive with existing models fine-tuned on each specific benchmark. We release OPT-IML at both scales, together with the OPT-IML Bench evaluation framework.

PaperPDFCode

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

Code

tanyuqian/cappy mentioned on GitHubjax 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

Language ModelingLanguage ModellingMeta-LearningNatural Language InferenceQuestion Answering

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Natural Language Inference RTE OPT-IML 175B Accuracy 84.8% #26 of 90 Archive leaderboard report
Natural Language Inference RTE OPT-IML 30B Accuracy 83.8% #31 of 90 Archive leaderboard report
Natural Language Inference RTE OPT-IML 1.3B Accuracy 66.8% #63 of 90 Archive leaderboard report
Natural Language Inference RTE OPT 175B Accuracy 60.3% #72 of 90 Archive leaderboard report
Natural Language Inference RTE OPT 30B Accuracy 58.1% #76 of 90 Archive leaderboard report
Natural Language Inference RTE OPT 1.3B Accuracy 54.2% #85 of 90 Archive leaderboard report
Question Answering BoolQ OPT-IML 175B Accuracy 71.4 #42 of 65 Archive leaderboard report
Question Answering BoolQ OPT-IML 30B Accuracy 66.9 #45 of 65 Archive leaderboard report
Question Answering BoolQ OPT 30B (0-shot) Accuracy 64 #49 of 65 Archive leaderboard report
Question Answering BoolQ OPT-IML 1.3B (0-shot) Accuracy 61.5 #53 of 65 Archive leaderboard report
Question Answering BoolQ OPT 1.3B (zero-shot) Accuracy 60.5 #56 of 65 Archive leaderboard report
Question Answering BoolQ OPT 175B Accuracy 60.1 #58 of 65 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

Introduced by this paper: OPT-IML

OPTOPT-IML

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