Papers › Filter, Then Reweight: Rethinking Optimization Granularity in On-Policy Distillation

Filter, Then Reweight: Rethinking Optimization Granularity in On-Policy Distillation

1 Jun 2026arXiv:2606.02684added by Syntology

Yuying Li, Leqi Zheng, Yongzi Yu, Wenrui Zhou, Xuchang Zhong, Xing Hu, Jing Jin, Hangjie Yuan, Tao Feng

Title, abstract, authors and date from arXiv's metadata (CC0); this paper is not in the Papers with Code archive (frozen 2025-07-28).

On-Policy distillation (OPD) in large language models is shifting from full-trace KL supervision toward more selective training paradigms. Recent OPD methods increasingly focus on selecting which trajectories to learn from, which tokens are most informative, and which supervision signals are most reliable. Motivated by this trend, we rethink optimization granularity of OPD and propose \fireicon FiRe-OPD (Filter, then Reweight), which jointly adjusts supervision signals at both trajectory and token levels. In details, FiRe-OPD first filters trajectories to remove low-quality rollout samples, and then applies soft reweighting within the retained trajectories to emphasize informative tokens. Compared with hard token selection, FiRe-OPD leverages a soft-weighting mechanism to effectively mitigate information loss and enhance optimization stability, thereby achieving finer-grained OPD optimization. We validate the effectiveness of FiRe-OPD across strong-to-weak, single-teacher, and multi-teacher settings, and demonstrate its superiority over recent token-level OPD methods ( (e.g., +6.25 on AIME 2024 in strong-to-weak, +18.81 on Miner in multi-teacher). Our code is available at https://github.com/YuYingLi0/FiRe-OPD.

PaperPDFCode 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 12 code samples harvested from 1 repository linked to this paper: 12 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 0 violated, 3 with no contract checked; 8 where Syntology's instrument failed) · 0 unverified. Where Syntology's instrument failed: 6 where our contract draft was wrong; 2 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: found in paper text by Syntology: 12 samples from 1 repository, 12 ran (4 with no instrument failure; 8 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 between 2026-09-22 and 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

YuYingLi0/FiRe-OPD found in paper text by SyntologySyntology: 12 samples, 12 ran (8 instrument-failed) · 0 unverified report

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

12 samples harvested: 12 ran (of which 0 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 0 violated, 3 with no contract checked; 8 where Syntology's instrument failed) · 0 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
6ran · our draft was wrong
2ran; the contract check could not run
3ran

Licence: 12 of the 12 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 YuYingLi0/FiRe-OPD. “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.

YuYingLi0/FiRe-OPD found in paper text by Syntology 12 samples, 12 ran (8 instrument-failed) · 0 unverified
apply_chat_template in 44 papers math_eval/eval_math.py ran no licence file found · pointer only · 6eceb1b9ba215e62 · report fixture: signature toker:MockTokenizer/0/?, messages:list/1/dict, chat_template:MockChatTemplate/0/?, enable_thinking:bool/0/bool · returned str · level recorded 2026-09-24
compute_ce_dpo_loss_rm in 48 papers verl/recipe/prime/prime_core_algos.py ran no licence file found · pointer only · f220595adfc5e83c · report fixture: signature token_level_scores:Tensor/2/float, acc:Tensor/1/float, response_mask:Tensor/2/float, beta:float/0/float · returned Tensor · level recorded 2026-09-24
compute_detach_dpo_loss_rm in 49 papers verl/recipe/prime/prime_core_algos.py ran · our draft was wrong no licence file found · pointer only · 4d91e4383716e42c · report fixture: signature token_level_scores:Tensor/2/float, acc:Tensor/1/float, Q_bc:list/1/Tensor, acc_bc:Tensor/2/float, response_mask:Tensor/2/float, beta:float/0/float, bon_mode:str/0/str · returned Tensor · level recorded 2026-09-24
compute_online_dpo_loss in 39 papers verl/recipe/spin/core_algos.py ran no licence file found · pointer only · f19825983a523003 · report fixture: signature policy_chosen_logps:Tensor/1/float, policy_rejected_logps:Tensor/1/float, reference_chosen_logps:Tensor/1/float, reference_rejected_logps:Tensor/1/float, beta:float/0/float, label_smoothing:float/0/float, loss_type:str/0/str, reference_free:bool/0/bool · returned Tensor · level recorded 2026-09-24
compute_onlinedpo_pref in 39 papers verl/recipe/spin/core_algos.py ran · our draft was wrong no licence file found · pointer only · 41b6e81dc3ecba79 · report fixture: signature token_level_rewards:Tensor/2/float, response_mask:Tensor/2/float · returned Tensor · level recorded 2026-09-24
compute_score_data_source in 14 papers verl/recipe/open_math_reasoning/compute_score.py ran; the contract check could not run (DEP_MISSING) no licence file found · pointer only · 05a86ccb0b3876dd · report fixture: signature data_source:str/0/str, response:str/0/str, ground_truth:str/0/str · returned float · level recorded 2026-09-24
generate_config_from_args in 68 papers verl/verl/model_merger/base_model_merger.py ran · our draft was wrong no licence file found · pointer only · 8896f4e2bafe8abc · report fixture: signature args:Namespace/0/? · returned ModelMergerConfig · level recorded 2026-09-24
get_dynamic_pipeline_shards in 53 papers verl/verl/model_merger/megatron_model_merger.py ran · honoured contract fingerprinted no licence file found · pointer only · 94209884dcb2308e · report fixture: signature layer_num:int/0/int, pp_size:int/0/int · returned list · level recorded 2026-09-24
get_kl_controller in 53 papers verl/recipe/spin/core_algos.py ran · our draft was wrong no licence file found · pointer only · 757d109d90121142 · report fixture: signature kl_ctrl:KLControlConfig/0/? · returned AdaptiveKLController · level recorded 2026-09-24
last_boxed_only_string in 46 papers math_eval/eval_math.py ran · our draft was wrong fingerprinted no licence file found · pointer only · 90b5c896e5eaea5e · report fixture: signature string:str/0/str · returned str · level recorded 2026-09-22
remove_boxed in 48 papers math_eval/eval_math.py ran · our draft was wrong fingerprinted no licence file found · pointer only · 9a85fff3911bdfa2 · report fixture: signature s:str/0/str · returned str · level recorded 2026-09-22
reward_func in 49 papers verl/recipe/r1/reward_score.py ran; the contract check could not run (DEP_MISSING) · metamorphic tier: deterministic no licence file found · pointer only · 8a84f8cc921357e2 · report fixture: signature data_source:str/0/str, solution_str:str/0/str, ground_truth:str/0/str, extra_info:NoneType/0/? · returned Mock · level recorded 2026-09-22

Results from the paper

The Papers with Code archive ends with its 2025-07-28 snapshot. This paper's arXiv identifier, 2606.02684, was issued in June 2026, after that date, so the archive has no leaderboard rows for it.

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