Papers › KTO: Model Alignment as Prospect Theoretic Optimization

KTO: Model Alignment as Prospect Theoretic Optimization

2 Feb 2024arXiv:2402.01306archive 2025-07-28

Kawin Ethayarajh, Winnie Xu, Niklas Muennighoff, Dan Jurafsky, Douwe Kiela

Kahneman & Tversky's prospect theory tells us that humans perceive random variables in a biased but well-defined manner (1992); for example, humans are famously loss-averse. We show that objectives for aligning LLMs with human feedback implicitly incorporate many of these biases -- the success of these objectives (e.g., DPO) over cross-entropy minimization can partly be ascribed to them belonging to a family of loss functions that we call human-aware losses (HALOs). However, the utility functions these methods attribute to humans still differ from those in the prospect theory literature. Using a Kahneman-Tversky model of human utility, we propose a HALO that directly maximizes the utility of generations instead of maximizing the log-likelihood of preferences, as current methods do. We call this approach KTO, and it matches or exceeds the performance of preference-based methods at scales from 1B to 30B, despite only learning from a binary signal of whether an output is desirable. More broadly, our work suggests that there is no one HALO that is universally superior; the best loss depends on the inductive biases most appropriate for a given setting, an oft-overlooked consideration.

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 11 code samples harvested from 2 repositories linked to this paper: 5 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 2 where Syntology's instrument failed) · 6 unverified. Where Syntology's instrument failed: 2 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 9 samples from 1 repository, 5 ran (3 with no instrument failure; 2 where Syntology's instrument failed); community: 2 samples from 1 repository, 0 ran (0 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.

No sample is linked to this paper from 4 of its repositories that Syntology harvested for other papers:

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

contextualai/halos officialmentioned in papermentioned on GitHubpytorchApache-2.0Syntology: 9 samples, 5 ran (2 instrument-failed) · 4 unverified report
openbmb/eurus mentioned on GitHubpytorchApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Advancing LLM Reasoning Generalists with Preference Trees (arXiv:2404.02078; that paper's own run record: 18 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 2 honoured, 1 violated, 1 with no contract checked; 14 where Syntology's instrument failed) · 2 unverified); nothing checks that they implement this paper's method. report
reml-group/deliberation-on-priors mentioned on GitHubpytorchSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Reasoning on Graphs: Faithful and Interpretable Large Language Model Reasoning (arXiv:2310.01061; that paper's own run record: 0 ran · 1 unverified) and Deliberation on Priors: Trustworthy Reasoning of Large Language Models on Knowledge Graphs (arXiv:2505.15210; that paper's own run record: 5 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 2 honoured, 0 violated, 0 with no contract checked; 3 where Syntology's instrument failed) · 9 unverified); nothing checks that they implement this paper's method. report
tidedra/vl-rlhf mentioned on GitHubpytorchApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to ACL Anthology id 2025.naacl-long.447 (that paper's own run record: 1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 3 unverified); nothing checks that they implement this paper's method. report
uppaal/detox-edit mentioned on GitHubpytorchMITSyntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Model Editing as a Robust and Denoised variant of DPO: A Case Study on Toxicity (arXiv:2405.13967; that paper's own run record: 7 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 5 where Syntology's instrument failed) · 5 unverified); nothing checks that they implement this paper's method. 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.

A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.

Code Syntology ran Syntology

11 samples harvested: 5 ran (of which 0 constructed an object rather than computing a result; 3 with no instrument failure: 0 honoured, 0 violated, 3 with no contract checked; 2 where Syntology's instrument failed) · 6 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.

2ran · our draft was wrong
3ran
6unverified

Licence: 2 of the 11 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 2 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. “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.

contextualai/halos official repository 9 samples, 5 ran (2 instrument-failed) · 4 unverified
convert_to_binary_feedback train/label.py ran · our draft was wrong Apache-2.0 (permissive) · 65cbf1ea1a7a5e74 · report fixture: signature samples:list/1/dict, threshold:int/0/int · returned list · level recorded 2026-09-24
get_api_completion train/utils.py ran Apache-2.0 (permissive) · 9257a83ca14f9dd7 · report fixture: signature client:MockClient/0/?, system_prompt:str/0/str, user_prompt:str/0/str, model:str/0/str, temperature:float/0/float, max_retries:int/0/int, retry_delay:float/0/float · returned coroutine · level recorded 2026-09-24
get_feedback in 8 papers train/data.py ran Apache-2.0 (permissive) · 1ed0ab0ac3d9655a · report fixture: signature feedback_path:str/0/str, split:str/0/str · returned Dataset · level recorded 2026-09-24
pad_to_length in 18 papers train/utils.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 38bf6f7230fcc687 · report fixture: signature tensor:Tensor/2/float, length:int/0/int, pad_value:float/0/float, dim:int/0/int · returned Tensor · level recorded 2026-09-24
process_samples_with_api train/label.py ran Apache-2.0 (permissive) · 500478230027c795 · report fixture: signature samples:list/1/dict, client:callable/0/?, system_prompt:str/0/str, label_prompt:str/0/str, model:str/0/str, batch_size:int/0/int · returned coroutine · level recorded 2026-09-24
get_alpacaeval train/data.py unverified Apache-2.0 (permissive) · c54cf81b72eecb67 · report
get_base_model_state_dict_from_peft train/utils.py unverified Apache-2.0 (permissive) · 124288769111d62f · report
get_sampled_data train/data.py unverified Apache-2.0 (permissive) · 238fc17a8a274cea · report
process_batch_with_reward_model train/label.py unverified Apache-2.0 (permissive) · 9bf64659a7cf741d · report
jacklishufan/diffusion-kto community 2 samples, 0 ran · 2 unverified
parse_args in 355 papers train_kto_sd_v1.5.py unverified no licence file found · pointer only · 20f9d022f6c9a501 · report
tokenize_captions in 6 papers train_kto_sd_v1.5.py unverified no licence file found · pointer only · 28cae3111ef3900a · report

Tasks

Attributemodel

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

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