Papers › Compositional preference models for aligning LMs

Compositional preference models for aligning LMs

17 Oct 2023arXiv:2310.13011archive 2025-07-28

Dongyoung Go, Tomasz Korbak, Germán Kruszewski, Jos Rozen, Marc Dymetman

As language models (LMs) become more capable, it is increasingly important to align them with human preferences. However, the dominant paradigm for training Preference Models (PMs) for that purpose suffers from fundamental limitations, such as lack of transparency and scalability, along with susceptibility to overfitting the preference dataset. We propose Compositional Preference Models (CPMs), a novel PM framework that decomposes one global preference assessment into several interpretable features, obtains scalar scores for these features from a prompted LM, and aggregates these scores using a logistic regression classifier. Through these simple steps, CPMs allow to control which properties of the preference data are used to train the preference model and to build it based on features that are believed to underlie the human preference judgment. Our experiments show that CPMs not only improve generalization and are more robust to overoptimization than standard PMs, but also that best-of-n samples obtained using CPMs tend to be preferred over samples obtained using conventional PMs. Overall, our approach demonstrates the benefits of endowing PMs with priors about which features determine human preferences while relying on LM capabilities to extract those features in a scalable and robust way.

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 13 code samples harvested from 1 repository linked to this paper: 11 ran (of which 0 constructed an object rather than computing a result; 11 with no instrument failure: 0 honoured, 0 violated, 11 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code; here there are none.

By repository: official repository: 13 samples from 1 repository, 11 ran (11 with no instrument failure; 0 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.

dongyoung-go/cpm officialmentioned in papermentioned on GitHubpytorchSyntology: 13 samples, 11 ran (0 instrument-failed) · 2 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

13 samples harvested: 11 ran (of which 0 constructed an object rather than computing a result; 11 with no instrument failure: 0 honoured, 0 violated, 11 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

11ran
2unverified

Licence: 13 of the 13 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 dongyoung-go/cpm. “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.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity.

dongyoung-go/cpm official repository 13 samples, 11 ran (0 instrument-failed) · 2 unverified
add_callback_features feature_extract/reward_scaling_lll.py ran no licence file found · pointer only · 25a2fa5e6d781899 · report fixture: signature scored_samples:dict/0/?, feature_callbacks:dict/0/? · returned dict · level recorded 2026-09-24
features_values feature_extract/reward_scaling_lll.py ran no licence file found · pointer only · 64c1c6dcbc84c35f · report fixture: signature scored_samples:DataFrame/0/?, feature_names:list/1/str · returned ndarray · level recorded 2026-09-24
find_score mle-train/claude_eval.py ran no licence file found · pointer only · 61ba2e9faf99f486 · report fixture: signature score_list:list/1/str · returned list · level recorded 2026-09-24
is_single_round feature_extract/sample_generate.py ran no licence file found · pointer only · 0c518b646b15ecf4 · report fixture: signature element:dict/0/? · returned bool · level recorded 2026-09-24
parse_ds mle-train/common.py ran no licence file found · pointer only · 09c62c73831a3da2 · report fixture: signature ds:list/1/dict, feature_names:list/1/str, context:str/0/str, response_prefix:str/0/str, score_prefix:str/0/str, options:list/1/str, feature_callbacks:dict/0/? · returned tuple · level recorded 2026-09-24
parse_features_ds mle-train/common.py ran no licence file found · pointer only · 9dc7278bf125c774 · report fixture: signature ds:list/1/dict, feature_names:list/1/str, context_key:str/0/str, response_key:str/0/str, score_key:str/0/str, feature_callbacks:dict/0/? · returned tuple · level recorded 2026-09-24
preferences_as_str mle-train/common.py ran no licence file found · pointer only · a641066a5fab4c04 · report fixture: signature preferences:list/1/int · returned list · level recorded 2026-09-24
preprocess_hh_dataset reward_model/rm_dataset.py ran no licence file found · pointer only · ceca6c9a0952d84c · report fixture: signature example:dict/0/?, prepend_mode:str/0/str · returned dict · level recorded 2026-09-24
preprocess_output reward_model/RM_inference_upload.py ran no licence file found · pointer only · 2a1243ce1a23e284 · report fixture: signature model_pickle:dict/0/? · returned dict · level recorded 2026-09-24
preprocess_shp_dataset reward_model/rm_dataset.py ran no licence file found · pointer only · 9427415e5a3dcdf6 · report fixture: signature example:dict/0/?, prepend_mode:str/0/str · returned dict · level recorded 2026-09-24
standardize in 47 papers mle-train/bon_candidate.py ran fingerprinted no licence file found · pointer only · 790643e0394e8970 · report fixture: signature data:Tensor/2/float · returned Tensor · level recorded 2026-09-24
load_and_add_score mle-train/bon_candidate.py unverified no licence file found · pointer only · 86ee092ffb48abf3 · report
load_scored_samples feature_extract/reward_scaling_lll.py unverified no licence file found · pointer only · fdea1016316c5a9a · report

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

ALIGNLogistic Regression

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