Papers › ReFT: Reasoning with Reinforced Fine-Tuning

ReFT: Reasoning with Reinforced Fine-Tuning

17 Jan 2024arXiv:2401.08967archive 2025-07-28

Trung Quoc Luong, Xinbo Zhang, Zhanming Jie, Peng Sun, Xiaoran Jin, Hang Li

One way to enhance the reasoning capability of Large Language Models (LLMs) is to conduct Supervised Fine-Tuning (SFT) using Chain-of-Thought (CoT) annotations. This approach does not show sufficiently strong generalization ability, however, because the training only relies on the given CoT data. In math problem-solving, for example, there is usually only one annotated reasoning path for each question in the training data. Intuitively, it would be better for the algorithm to learn from multiple annotated reasoning paths given a question. To address this issue, we propose a simple yet effective approach called Reinforced Fine-Tuning (ReFT) to enhance the generalizability of learning LLMs for reasoning, with math problem-solving as an example. ReFT first warmups the model with SFT, and then employs on-line reinforcement learning, specifically the PPO algorithm in this paper, to further fine-tune the model, where an abundance of reasoning paths are automatically sampled given the question and the rewards are naturally derived from the ground-truth answers. Extensive experiments on GSM8K, MathQA, and SVAMP datasets show that ReFT significantly outperforms SFT, and the performance can be potentially further boosted by combining inference-time strategies such as majority voting and re-ranking. Note that ReFT obtains the improvement by learning from the same training questions as SFT, without relying on extra or augmented training questions. This indicates a superior generalization ability for ReFT.

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: 5 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 1 honoured, 1 violated, 3 with no contract checked; 0 where Syntology's instrument failed) · 9 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code; here there are none.

By repository: official repository: 14 samples from 1 repository, 5 ran (5 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 2026-09-25 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

lqtrung1998/mwp_reft officialmentioned in papermentioned on GitHubpytorchApache-2.0Syntology: 14 samples, 5 ran (0 instrument-failed) · 9 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: 5 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 1 honoured, 1 violated, 3 with no contract checked; 0 where Syntology's instrument failed) · 9 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 · honoured contract
1ran · violated contract
3ran
9unverified

Licence: 13 of the 14 samples are pointer only, meaning Syntology does not serve that copy's text. For 13 of them, this paper's copy is under a permissive licence and cleared, but the stored copy of the same code that Syntology serves is not; the licence label's tooltip gives Syntology's recorded reason. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from lqtrung1998/mwp_reft. “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.

lqtrung1998/mwp_reft official repository 14 samples, 5 ran (0 instrument-failed) · 9 unverified
collate_fn in 500 papers sampling.py ran Apache-2.0 (permissive) here; the stored copy is pointer only · 2bb5bd3526f55ecb · report fixture: signature features:list/1/dict, tokenizer:MockTokenizer/0/? · returned dict · level recorded 2026-09-25
floatify in 9 papers src/utils.py ran · honoured contract fingerprinted Apache-2.0 (permissive) here; the stored copy is pointer only · f21a5f0d5bae0c14 · report fixture: signature s:str/0/str · returned float · level recorded 2026-09-25
is_numeric in 28 papers src/utils.py ran · violated contract fingerprinted Apache-2.0 (permissive) here; the stored copy is pointer only · c3b86a3bceec86cd · report fixture: signature value:int/0/int · returned bool · level recorded 2026-09-25
process_code in 7 papers src/python_engine.py ran fingerprinted Apache-2.0 (permissive) here; the stored copy is pointer only · 8b9f60df8d2c8a82 · report fixture: signature code_gen:str/0/str, truncate_first_return:bool/0/bool · returned str · level recorded 2026-09-25
tokenize_fn in 9 papers sampling.py ran Apache-2.0 (permissive) here; the stored copy is pointer only · 951fe709737c50b9 · report fixture: signature examples:dict/0/?, tokenizer:MinimalTokenizer/0/?, max_length:int/0/int, src_name:str/0/str, engine:str/0/str · returned dict · level recorded 2026-09-25
do_gather src/utils.py unverified Apache-2.0 (permissive) here; the stored copy is pointer only · 32851be78f5ecd4c · report
eval_code src/python_engine.py unverified Apache-2.0 (permissive) here; the stored copy is pointer only · 7fcae9f5e120bafe · report
evaluate_rerank train_reward_model.py unverified Apache-2.0 (permissive) here; the stored copy is pointer only · c8164df09b7d7a68 · report
prepare_datasets_and_data_loaders train_reward_model.py unverified Apache-2.0 (permissive) here; the stored copy is pointer only · 4d434f6e5fd591b5 · report
prepare_datasets_and_data_loaders train_sl_model.py unverified Apache-2.0 (permissive) here; the stored copy is pointer only · 9f0489fe01df9539 · report
prepare_datasets_and_data_loaders train_rl_reft.py unverified Apache-2.0 (permissive) here; the stored copy is pointer only · 449e6a4081f8d019 · report
prepare_datasets_and_data_loaders train_rl_sl.py unverified Apache-2.0 (permissive) here; the stored copy is pointer only · 01e9a5c8f882c073 · report
prepare_datasets_and_data_loaders train_sft_model.py unverified Apache-2.0 (permissive) · d0eff44d0f627c24 · report
train_one_epoch in 50 papers train_reward_model.py unverified Apache-2.0 (permissive) here; the stored copy is pointer only · d3b8681876990bfa · report

Tasks

GSM8KMathRe-Ranking

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Entropy RegularizationPPOSFT

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