SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2605.09548 · 2026

Crosslingual On-Policy Self-Distillation for Multilingual Reasoning

Michael Hedderich, Hinrich Schütze, Yihong Liu, Raoyuan Zhao

arXiv · PDF · Open in the Atlas

Code that ran

8 of 8 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 1 violated it and 0 have no contract verdict; 7 failures of our instrument; 0 no recorded run), measured 2026-11-22; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2605.09548.json · badge: /api/ran/2605.09548.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
cisnlp/COPSD canonical (classifier label)
classifier basis: census
8 ran · 0 no recorded run
FunctionStatusWhere it lives
add_target_language Ran
ran; our contract draft was wrong, not the code
cisnlp/COPSD/multilingual_opsd_train.py
pointer only (licence: NONE) · get_code("39cb5833ed0cc28b")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, cisnlp/COPSD when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

build_qwen3_prompt Ran
ran; our contract draft was wrong, not the code
cisnlp/COPSD/polymath_eval/evaluate_math.py
pointer only (licence: NONE) · get_code("24e9924c0bfa561d")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, cisnlp/COPSD when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

extract_boxed_answer Ran
ran; our contract draft was wrong, not the code
cisnlp/COPSD/multilingual_grpo_train.py
pointer only (licence: NONE) · get_code("90054f5de2b3ede3")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, cisnlp/COPSD when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

extract_boxed_answer Ran
ran; our contract draft was wrong, not the code
cisnlp/COPSD/african_langs_eval/evaluate_math.py
pointer only (licence: NONE) · get_code("48da3f1fbb868b57")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, cisnlp/COPSD when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

has_target_translation Ran
ran, violated the contract we drafted
cisnlp/COPSD/multilingual_opsd_train.py
pointer only (licence: NONE) · get_code("0ff19ae07f98c505")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, cisnlp/COPSD when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

normalize_polymath_language Ran
ran; our contract draft was wrong, not the code
cisnlp/COPSD/african_langs_eval/evaluate_math.py
pointer only (licence: NONE) · get_code("1ae94b058b051ae7")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, cisnlp/COPSD when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

resolve_translated_data_path Ran
ran; our contract draft was wrong, not the code
cisnlp/COPSD/multilingual_grpo_train.py
pointer only (licence: NONE) · get_code("cfd534241c7065f0")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, cisnlp/COPSD when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

resolve_translated_data_path Ran
ran; our contract draft was wrong, not the code
cisnlp/COPSD/multilingual_opsd_train.py
pointer only (licence: NONE) · get_code("f04f77b0981bd2e1")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, cisnlp/COPSD when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 distinct functions with an identity, 1 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 1 later) and 7 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
extract_boxed_answer 3 other papers
1 dated earlier, 2 later
2601.18734 (2026-01)
Self-Distilled Reasoner: On-Policy Self-Distillation for Large Language Models
2606.23104 (2026-06)
ReNIO: Reweighting Negative Trajectory Importance for LLM On-Policy Distillation
2608.27448 (2026-08)
TTPO: Test-Time Policy Optimization

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Large language models (LLMs) have achieved remarkable progress in mathematical reasoning, but this ability is not equally accessible across languages. Especially low-resource languages exhibit much lower reasoning performance. To address this, we propose Crosslingual On-Policy Self-Distillation (COPSD), which transfers a model's own high-resource reasoning behavior to low-resource languages. COPSD uses the same model as student and teacher: the student sees only the low-resource problem, while the teacher receives privileged crosslingual context, including the problem translation and reference solution in English. Training minimizes full-distribution token-level divergence on the student's own rollouts, providing dense supervision while avoiding the sparsity and instability of outcome-only reinforcement learning (RL). Experiments on 17 lowresource African languages show that COPSD consistently improves low-resource mathematical reasoning across model sizes and substantially outperforms Group Relative Policy Optimization (GRPO). Further analyses show that COPSD improves answer-format adherence, strengthens test-time scaling, and generalizes to harder multilingual reasoning benchmarks, with especially large gains for lower-resource languages. We make our code and data available at https://github.com/cisnlp/COPSD.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2605.09548")
get_code_for_paper("2605.09548")
have("2605.09548")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2605.09548.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2605.09548.svg)](https://syntology.ai/paper/2605.09548)

Connect an agent — have() is free.