Papers › Towards Modular LLMs by Building and Reusing a Library of LoRAs

Towards Modular LLMs by Building and Reusing a Library of LoRAs

18 May 2024arXiv:2405.11157archive 2025-07-28

Oleksiy Ostapenko, Zhan Su, Edoardo Maria Ponti, Laurent Charlin, Nicolas Le Roux, Matheus Pereira, Lucas Caccia, Alessandro Sordoni

The growing number of parameter-efficient adaptations of a base large language model (LLM) calls for studying whether we can reuse such trained adapters to improve performance for new tasks. We study how to best build a library of adapters given multi-task data and devise techniques for both zero-shot and supervised task generalization through routing in such library. We benchmark existing approaches to build this library and introduce model-based clustering, MBC, a method that groups tasks based on the similarity of their adapter parameters, indirectly optimizing for transfer across the multi-task dataset. To re-use the library, we present a novel zero-shot routing mechanism, Arrow, which enables dynamic selection of the most relevant adapters for new inputs without the need for retraining. We experiment with several LLMs, such as Phi-2 and Mistral, on a wide array of held-out tasks, verifying that MBC-based adapters and Arrow routing lead to superior generalization to new tasks. We make steps towards creating modular, adaptable LLMs that can match or outperform traditional joint training.

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

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

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 14 code samples harvested from 1 repository linked to this paper: 13 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 1 violated, 6 with no contract checked; 6 where Syntology's instrument failed) · 1 unverified. Where Syntology's instrument failed: 6 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: community (archive-listed): 14 samples from 1 repository, 13 ran (7 with no instrument failure; 6 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written on 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

microsoft/mttl mentioned on GitHubpytorchMITSyntology: 14 samples, 13 ran (6 instrument-failed) · 1 unverified 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

14 samples harvested: 13 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 1 violated, 6 with no contract checked; 6 where Syntology's instrument failed) · 1 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · violated contract
6ran · our draft was wrong
6ran
1unverified

Licence: 0 of the 14 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 microsoft/mttl. “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. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (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 are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. 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 six 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; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. 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.

microsoft/mttl community (archive-listed) 14 samples, 13 ran (6 instrument-failed) · 1 unverified
augment_few_shot_task mttl/datamodule/mt_seq_to_seq_module.py ran MIT (permissive) · 9607e74609d06a5c · report fixture: signature dataset:MinimalDataset/0/?, num_samples:NoneType/0/?, few_shots:dict/0/?, tokenizer:callable/0/?, max_input_length:int/0/int, seed:int/0/int, modify_task_source:bool/0/bool · returned MinimalDataset · level recorded 2026-09-24
augment_prompts mttl/datamodule/mmlu_data_module.py ran · our draft was wrong MIT (permissive) · 91c46d2158999132 · report fixture: signature dataset:MockDataset/0/? · returned dict · level recorded 2026-09-24
completion_template mttl/datamodule/humaneval_module.py ran · our draft was wrong MIT (permissive) · 90b13d358f062991 · report fixture: signature example:dict/0/? · returned dict · level recorded 2026-09-24
completion_template mttl/datamodule/mbpp_datamodule.py ran MIT (permissive) · 3d1190e58b358dda · report fixture: signature for_generation:bool/0/bool, example:dict/0/? · returned dict · level recorded 2026-09-24
create_config_class_from_args mttl/arguments.py ran MIT (permissive) · 0a70817429b93f25 · report fixture: signature config_class:callable/0/?, args:ArgsNamespace/0/? · returned TestConfig · level recorded 2026-09-24
detect_indentation mttl/datamodule/mbpp_datamodule.py ran · our draft was wrong fingerprinted MIT (permissive) · 801642274b17ca93 · report fixture: signature func:str/0/str · returned str · level recorded 2026-09-24
instruct_template mttl/datamodule/humaneval_module.py ran · our draft was wrong MIT (permissive) · ef9fcdcd4db23c46 · report fixture: signature example:dict/0/? · returned dict · level recorded 2026-09-24
instruct_template mttl/datamodule/mbpp_datamodule.py ran · our draft was wrong MIT (permissive) · d549f96aaac26455 · report fixture: signature example:dict/0/? · returned dict · level recorded 2026-09-24
is_phi2_eval_task mttl/datamodule/mt_seq_to_seq_module.py ran · violated contract MIT (permissive) · ea35b5a896e6de9d · report fixture: signature task:str/0/str · returned bool · level recorded 2026-09-24
maybe_filter_hf_dataset_by_key mttl/datamodule/clip_data_module.py ran MIT (permissive) · 8df019af71f4ed70 · report fixture: signature dataset:dict/0/?, key:str/0/str, task_names:str/0/str, n_proc:int/0/int · returned tuple · level recorded 2026-09-24
permute_options mttl/datamodule/mmlu_data_module.py ran · our draft was wrong MIT (permissive) · 0a3ee7f35121b4ea · report fixture: signature dataset:MockDataset/0/? · returned MockDataset · level recorded 2026-09-24
subsample_dst mttl/datamodule/base.py ran MIT (permissive) · 6b3aac9749e83a24 · report fixture: signature dataset:SimpleDataset/0/?, subsample:int/0/int, rng:Generator/0/? · returned Subset · level recorded 2026-09-24
take_n_examples_per_task mttl/datamodule/base.py ran MIT (permissive) · f7e24d6e9c199261 · report fixture: signature task_names:list/1/str, n:int/0/int, rng:RandomState/0/? · returned list · level recorded 2026-09-24
augment_few_shot mttl/datamodule/mt_seq_to_seq_module.py unverified MIT (permissive) · 6773eed2108a3c15 · report

Tasks

Language ModelingLanguage ModellingLarge Language Model

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

AdapterBASE

1 archive method tag without a method page not shown.

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