Papers › Free Process Rewards without Process Labels

Free Process Rewards without Process Labels

2 Dec 2024arXiv:2412.01981archive 2025-07-28

Lifan Yuan, Wendi Li, Huayu Chen, Ganqu Cui, Ning Ding, Kaiyan Zhang, BoWen Zhou, Zhiyuan Liu, Hao Peng

Different from its counterpart outcome reward models (ORMs), which evaluate the entire responses, a process reward model (PRM) scores a reasoning trajectory step by step, providing denser and more fine grained rewards. However, training a PRM requires labels annotated at every intermediate step, presenting significant challenges for both manual and automatic data collection. This paper aims to address this challenge. Both theoretically and empirically, we show that an implicit PRM can be obtained at no additional cost, by simply training an ORM on the cheaper response-level labels. The only assumption is to parameterize the outcome reward as the log-likelihood ratios of the policy and reference models, which can be optimized regardless of the specific choice of loss objectives. In experiments, we instantiate our implicit PRMs with various objectives and evaluate their performance on MATH. We show that our implicit PRM outperforms a strong MCTS-based baseline á la Math-Shepherd using less than 1/38 of the training data. Its performance can be further improved with majority voting. We further find that scaling up instructions and responses benefits our implicit PRM, and the latter brings a larger gain. Particularly, we find that our implicit PRM, when instantiated with the cross-entropy (CE) loss, is more data-efficient and can keep improving generation models even when trained with only one response per instruction, the setup that suffers from extreme data scarcity and imbalance. Further, instructions should be relevant to downstream tasks while the diversity of responses does not bring gains. Surprisingly, training on extra Math-Shepherd step labels brings no further improvements to our implicit PRM trained on only outcome data. We hope that our work will encourage a rethinking of PRM training approaches and contribute to making training PRMs more accessible.

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 22 code samples harvested from 1 repository linked to this paper: 14 ran (of which 0 constructed an object rather than computing a result; 8 with no instrument failure: 5 honoured, 0 violated, 3 with no contract checked; 6 where Syntology's instrument failed) · 8 unverified. Where Syntology's instrument failed: 6 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 22 samples from 1 repository, 14 ran (8 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.

No sample is linked to this paper from 1 of its repositories that Syntology harvested for other papers:

Levels written between 2026-09-22 and 2026-09-26 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

lifan-yuan/implicitprm officialmentioned in papermentioned on GitHubpytorchApache-2.0Syntology: 22 samples, 14 ran (6 instrument-failed) · 8 unverified report
prime-rl/prime mentioned on GitHubpytorchApache-2.0Syntology: no sample linked to this paper (harvested for another paper). Syntology's graph links these samples to Process Reinforcement through Implicit Rewards (arXiv:2502.01456; that paper's own run record: 14 ran (of which 0 constructed an object rather than computing a result; 10 with no instrument failure: 2 honoured, 0 violated, 8 with no contract checked; 4 where Syntology's instrument failed) · 3 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

22 samples harvested: 14 ran (of which 0 constructed an object rather than computing a result; 8 with no instrument failure: 5 honoured, 0 violated, 3 with no contract checked; 6 where Syntology's instrument failed) · 8 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.

5ran · honoured contract
6ran · our draft was wrong
3ran
8unverified

Licence: 1 of the 22 samples is pointer only, meaning Syntology does not serve that copy's text. For that sample, 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 lifan-yuan/implicitprm. “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. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

lifan-yuan/implicitprm official repository 22 samples, 14 ran (6 instrument-failed) · 8 unverified
clean_units eval/math_utils/eval_theoremqa.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 52095429576ac98d · report fixture: signature pred_str:str/0/str · returned str · level recorded 2026-09-25
compute_approx_kl in 20 papers train/openrlhf/models/utils.py ran fingerprinted Apache-2.0 (permissive) · b507127549cdea93 · report fixture: signature log_probs:Tensor/2/float, log_probs_base:Tensor/2/float, action_mask:Tensor/2/float · returned Tensor · level recorded 2026-09-25
compute_reward in 26 papers train/openrlhf/models/utils.py ran Apache-2.0 (permissive) · 38b7cb8077e28347 · report fixture: signature r:Tensor/1/float, kl_coef:float/0/float, log_probs:Tensor/2/float, log_probs_base:Tensor/2/float, action_mask:Tensor/2/float, num_actions:NoneType/0/? · returned tuple · level recorded 2026-09-25
extract_answer_string_2 eval/math_utils/normalizer.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · 906aa9a1e739f957 · report fixture: signature answer_str:str/0/str · returned str · level recorded 2026-09-25
find_threshold_sigmoid eval/processbench.py ran · honoured contract Apache-2.0 (permissive) · bc171287029c3465 · report fixture: signature data:list/1/dict, num_thresholds:int/0/int · returned tuple · level recorded 2026-09-25
floatify in 9 papers eval/math_utils/eval_theoremqa.py ran · honoured contract fingerprinted Apache-2.0 (permissive) · f8cd262700389f42 · report fixture: signature num:str/0/str · returned float · level recorded 2026-09-25
get_raw_data in 9 papers eval/prm_eval_utils.py ran Apache-2.0 (permissive) · 1d78ad671b732515 · report fixture: signature dataset_name:str/0/str, process_inst:bool/0/bool · returned tuple · level recorded 2026-09-25
handle_base in 6 papers eval/math_utils/grader.py ran · honoured contract fingerprinted Apache-2.0 (permissive) · bb175bff1060d928 · report fixture: signature x:str/0/str · returned int · level recorded 2026-09-26
is_digit in 33 papers eval/math_utils/grader.py ran · honoured contract fingerprinted Apache-2.0 (permissive) · 2a180d3156997924 · report fixture: signature s:str/0/str · returned tuple · level recorded 2026-09-26
log_probs_from_logits in 18 papers train/openrlhf/models/utils.py ran · our draft was wrong Apache-2.0 (permissive) · 959df7824bf09cc1 · report fixture: signature logits:Tensor/2/float, labels:Tensor/1/int · returned Tensor · level recorded 2026-09-22
normalize in 638 papers
census: 70 implementations from 82 papers, 51 distinct outputs across 11 buckets, one shared input per bucket
eval/math_utils/grader.py ran · our draft was wrong Apache-2.0 (permissive) · 960059cee2753e3f · report fixture: signature answer:str/0/str, pi:float/0/float · returned str · level recorded 2026-09-26
normalize_answer in 304 papers eval/math_utils/math_normalize.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) here; the stored copy is pointer only · 101b1e54fcdadf90 · report fixture: signature answer:str/0/str · returned str · level recorded 2026-09-24
normalize_answer_string eval/math_utils/normalizer.py ran · our draft was wrong fingerprinted Apache-2.0 (permissive) · c431597b4cd16dfc · report fixture: signature expr:str/0/str · returned str · level recorded 2026-09-25
sigmoid in 216 papers
census: 36 implementations from 56 papers, 7 distinct outputs across 3 buckets, one shared input per bucket
eval/processbench.py ran · honoured contract fingerprinted Apache-2.0 (permissive) · 1c192362b31daf5b · report fixture: signature x:ndarray/2/float · returned ndarray · level recorded 2026-09-25
compute_metrics in 208 papers eval/bon_eval.py unverified Apache-2.0 (permissive) · ad31a6c133b0728c · report
convert_ring_attn_params in 12 papers train/openrlhf/models/ring_attn_utils.py unverified Apache-2.0 (permissive) · d66f259aaf6cd9c7 · report
extract_attributes_from_name eval/math_utils/normalizer.py unverified Apache-2.0 (permissive) · 3dd56c766e1b4ca8 · report
get_tokenizer in 156 papers eval/prm_eval_utils.py unverified Apache-2.0 (permissive) · 7032a4070687fc83 · report
reset_ring_attn_position_ids in 16 papers train/openrlhf/models/ring_attn_utils.py unverified Apache-2.0 (permissive) · e93117d99aa6b210 · report
set_special_token_ids eval/prm_eval_utils.py unverified Apache-2.0 (permissive) · fb5dce3b9cdca298 · report
setup_accelerator_and_models eval/processbench.py unverified Apache-2.0 (permissive) · 5a3cf1b9cf6bcb3c · report
within_eps in 7 papers eval/math_utils/eval_theoremqa.py unverified Apache-2.0 (permissive) · 98b05a45dfc53b99 · report

Tasks

Math

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