Papers › Exploring the Benefits of Training Expert Language Models over Instruction Tuning

Exploring the Benefits of Training Expert Language Models over Instruction Tuning

7 Feb 2023arXiv:2302.03202archive 2025-07-28

Joel Jang, Seungone Kim, Seonghyeon Ye, Doyoung Kim, Lajanugen Logeswaran, Moontae Lee, Kyungjae Lee, Minjoon Seo

Recently, Language Models (LMs) instruction-tuned on multiple tasks, also known as multitask-prompted fine-tuning (MT), have shown the capability to generalize to unseen tasks. Previous work has shown that scaling the number of training tasks is the key component in making stronger MT LMs. In this work, we report an unexpected finding that an expert LM fine-tuned on just a single task can outperform an MT LM trained with 300+ different tasks on 11 different unseen datasets and on 13 datasets of the BIG-bench benchmark by a mean accuracy of 3.20% and 1.29%, respectively. This finding casts doubt on the previously held belief that simply scaling the number of tasks makes stronger MT LMs. Leveraging this finding, we further show that this distributed approach of training a separate expert LM per training task instead of a single MT LM for zero-shot inference possesses many benefits including (1) avoiding negative task transfer that often occurs during instruction tuning, (2) being able to continually learn new tasks without having to re-train on previous tasks to avoid catastrophic forgetting, and (3) showing compositional capabilities when merging individual experts together. The code is available at https://github.com/joeljang/ELM.

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology: its repositories, every harvested function with whether it ran, its licence and the call to fetch it.

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

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2302.03202")

Code

Syntology Ran 3 of 3 code samples harvested from 1 repository linked to this paper; 0 have no recorded run. Of those that ran: 3 ran · our draft was wrong.

By repository: official repository: 2 samples from 1 repository, 2 ran; 1 identical to code first harvested elsewhere. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

joeljang/elm officialmentioned in papermentioned on GitHubpytorch report
joeljang/rlphf mentioned on GitHubpytorch 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

3 samples harvested; 3 ran; 0 honoured the contract we drafted; 0 have no recorded run. Read from Syntology's graph 2026-09-25; that is when this build read the record, not when the samples ran.

3ran · our draft was wrong

Licence: 3 of the 3 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from joeljang/elm. Some samples are identical code Syntology first harvested from another repository; for those, this paper's copy is not located and its licence is not recorded. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

clease_jinja joeljang/elm/retrieval/run_retrieval.py official repository ran · our draft was wrong fingerprinted no licence file found · pointer only · 9aa2aaf773117b75 · report
euclidean_distance joeljang/elm/retrieval/run_retrieval.py official repository ran · our draft was wrong fingerprinted no licence file found · pointer only · d77e7ade95e0c350 · report
run_command identical code first harvested elsewhere ran · our draft was wrong fingerprinted licence of this copy not recorded · 44e48dfbdecee3da · report

Tasks

Common Sense ReasoningCoreference ResolutionNatural Language InferenceQuestion AnsweringSentence CompletionWord Sense Disambiguation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Common Sense Reasoning WinoGrande RoE-3B Accuracy 61.60 #48 of 77 Archive leaderboard report
Coreference Resolution Winograd Schema Challenge RoE-3B Accuracy 62.21 #52 of 82 Archive leaderboard report
Natural Language Inference ANLI test RoE-3B A1 35.49 #15 of 25 Archive leaderboard report
Natural Language Inference ANLI test RoE-3B A2 34.64 #15 of 25 Archive leaderboard report
Natural Language Inference ANLI test RoE-3B A3 31.22 #15 of 25 Archive leaderboard report
Natural Language Inference RTE RoE-3B Accuracy 64.01 #66 of 90 Archive leaderboard report
Question Answering COPA RoE-3B Accuracy 79.25 #37 of 60 Archive leaderboard report
Question Answering StoryCloze RoE-3B Accuracy 86.33 #9 of 23 Archive leaderboard report
Sentence Completion HellaSwag RoE-3B Accuracy 34.6 #81 of 89 Archive leaderboard report
Word Sense Disambiguation Words in Context RoE-3B Accuracy 52.97 #25 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.

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