SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2406.18629 · 2024

Step-DPO: Step-wise Preference Optimization for Long-chain Reasoning of LLMs

arXiv · PDF · Open in the Atlas

Code that ran

11 of 12 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; 1 honoured the contract we drafted, 1 violated it and 1 has no contract verdict; 8 failures of our instrument; 1 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; 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/2406.18629.json · badge: /api/ran/2406.18629.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
dvlab-research/step-dpo canonical (classifier label)
classifier basis: census
11 ran · 1 no recorded run
FunctionStatusWhere it lives
batch_data Ran
ran; our contract draft was wrong, not the code
dvlab-research/step-dpo/eval_math.py
pointer only (licence: NONE) · get_code("66b3b5450182da3a")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, dvlab-research/step-dpo 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_answers Ran
ran; our contract draft was wrong, not the code
dvlab-research/step-dpo/evaluation/data_processing/answer_extraction.py
pointer only (licence: NONE); the stored copy get_code serves, from PKU-SEC-Lab/SPEX, is pointer only (licence: NONE) · get_code("6390dea763ce122e")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, dvlab-research/step-dpo 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_program Ran
ran; our contract draft was wrong, not the code
dvlab-research/step-dpo/evaluation/eval/eval_utils.py
pointer only (licence: NONE); the stored copy get_code serves, from ChengpengLi1003/CoRT, is pointer only (licence: NONE) · get_code("1d961161c71cc153")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, dvlab-research/step-dpo 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_program_output Ran
ran; our contract draft was wrong, not the code
dvlab-research/step-dpo/evaluation/data_processing/answer_extraction.py
pointer only (licence: NONE); the stored copy get_code serves, from PKU-SEC-Lab/SPEX, is pointer only (licence: NONE) · get_code("0772e5020f6b0849")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, dvlab-research/step-dpo 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.

generate_completions Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
dvlab-research/step-dpo/evaluation/eval/utils.py
pointer only (licence: NONE) · get_code("df28d2ed3962986b")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, dvlab-research/step-dpo 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.

get_next_word_predictions Ran
ran on a synthesized input
dvlab-research/step-dpo/evaluation/eval/utils.py
pointer only (licence: NONE) · get_code("19dce514835bd816")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, dvlab-research/step-dpo 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_numeric Ran
ran, honoured the contract we drafted
dvlab-research/step-dpo/evaluation/eval/ocwcourses_eval_utils.py
pointer only (licence: NONE) · get_code("a8456a1cf57a39d2")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, dvlab-research/step-dpo 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_symbolic_equation Ran
ran; our contract draft was wrong, not the code
dvlab-research/step-dpo/evaluation/eval/ocwcourses_eval_utils.py
pointer only (licence: NONE) · get_code("3bfb1033a74d23ba")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, dvlab-research/step-dpo 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.

numeric_equality Ran
ran, violated the contract we drafted
dvlab-research/step-dpo/evaluation/eval/ocwcourses_eval_utils.py
pointer only (licence: NONE) · get_code("4951ea903a7ffa37")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, dvlab-research/step-dpo 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.

parse_question Ran
ran; our contract draft was wrong, not the code
dvlab-research/step-dpo/evaluation/eval/eval_utils.py
pointer only (licence: NONE) · get_code("49e328786ffb3f8f")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, dvlab-research/step-dpo 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.

strip_string Ran
ran; our contract draft was wrong, not the code
dvlab-research/step-dpo/evaluation/data_processing/answer_extraction.py
pointer only (licence: NONE) · get_code("7f3eb00b0552d5c6")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, dvlab-research/step-dpo 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.

score_completions Not yet run
unverified (harvested, no recorded run)
dvlab-research/step-dpo/evaluation/eval/utils.py
pointer only (licence: NONE) · get_code("8c63844c66973f8d")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, dvlab-research/step-dpo 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. No invocation recorded.

Also present in other papers' code

Of 12 samples lifted from this paper's linked repositories, 12 carry a function identity and 0 do not. Of the 12 distinct functions with an identity, 9 are also present in other papers whose code we hold (6 with at least one paper dated earlier, 1 the same month, 8 later) and 3 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
batch_data 7 other papers
5 dated earlier, 2 the same month
2309.12284 (2023-09)
MetaMath: Bootstrap Your Own Mathematical Questions for Large Language Models
2404.10346 (2024-04)
Self-Explore: Enhancing Mathematical Reasoning in Language Models with Fine-grained Rewards
2405.06680 (2024-05)
Exploring the Compositional Deficiency of Large Language Models in Mathematical Reasoning Through Trap Problems
and 4 more not named here
extract_boxed_answers 7 other papers
1 dated earlier, 6 later
2402.03300 (2024-02)
DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
2025.acl-long.1506 (2025 (year only))
Subtle Errors in Reasoning: Preference Learning via Error-injected Self-editing
2502.12067 (2025-02)
TokenSkip: Controllable Chain-of-Thought Compression in LLMs
and 4 more not named here
strip_string 6 other papers
1 dated earlier, 5 later
2402.03300 (2024-02)
DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
2025.acl-long.1506 (2025 (year only))
Subtle Errors in Reasoning: Preference Learning via Error-injected Self-editing
2502.12067 (2025-02)
TokenSkip: Controllable Chain-of-Thought Compression in LLMs
and 3 more not named here
extract_program 5 other papers
2 dated earlier, 3 later
2309.17452 (2023-09)
ToRA: A Tool-Integrated Reasoning Agent for Mathematical Problem Solving
2403.13315 (2024-03)
PuzzleVQA: Diagnosing Multimodal Reasoning Challenges of Language Models with Abstract Visual Patterns
2502.12067 (2025-02)
TokenSkip: Controllable Chain-of-Thought Compression in LLMs
and 2 more not named here
extract_program_output 5 other papers
1 dated earlier, 4 later
2402.03300 (2024-02)
DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
2025.acl-long.1506 (2025 (year only))
Subtle Errors in Reasoning: Preference Learning via Error-injected Self-editing
2502.12067 (2025-02)
TokenSkip: Controllable Chain-of-Thought Compression in LLMs
and 2 more not named here
get_next_word_predictions 3 other papers
1 dated earlier, 2 later
2310.06830 (2023-10)
Lemur: Harmonizing Natural Language and Code for Language Agents
2502.12067 (2025-02)
TokenSkip: Controllable Chain-of-Thought Compression in LLMs
2506.20512 (2025-06)
OctoThinker: Mid-training Incentivizes Reinforcement Learning Scaling
parse_question 2 other papers
2 later
2502.12067 (2025-02)
TokenSkip: Controllable Chain-of-Thought Compression in LLMs
2506.20512 (2025-06)
OctoThinker: Mid-training Incentivizes Reinforcement Learning Scaling
score_completions 2 other papers
2 later
2502.12067 (2025-02)
TokenSkip: Controllable Chain-of-Thought Compression in LLMs
2506.20512 (2025-06)
OctoThinker: Mid-training Incentivizes Reinforcement Learning Scaling
numeric_equality 1 other paper
1 later
2506.20512 (2025-06)
OctoThinker: Mid-training Incentivizes Reinforcement Learning Scaling

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

Mathematical reasoning presents a significant challenge for Large Language Models (LLMs) due to the extensive and precise chain of reasoning required for accuracy. Ensuring the correctness of each reasoning step is critical. To address this, we aim to enhance the robustness and factuality of LLMs by learning from human feedback. However, Direct Preference Optimization (DPO) has shown limited benefits for long-chain mathematical reasoning, as models employing DPO struggle to identify detailed errors in incorrect answers. This limitation stems from a lack of fine-grained process supervision. We propose a simple, effective, and data-efficient method called Step-DPO, which treats individual reasoning steps as units for preference optimization rather than evaluating answers holistically. Additionally, we have developed a data construction pipeline for Step-DPO, enabling the creation of a high-quality dataset containing 10K step-wise preference pairs. We also observe that in DPO, self-generated data is more effective than data generated by humans or GPT-4, due to the latter's out-of-distribution nature. Our findings demonstrate that as few as 10K preference data pairs and fewer than 500 Step-DPO training steps can yield a nearly 3% gain in accuracy on MATH for models with over 70B parameters. Notably, Step-DPO, when applied to Qwen2-72B-Instruct, achieves scores of 70.8% and 94.0% on the test sets of MATH and GSM8K, respectively, surpassing a series of closed-source models, including GPT-4-1106, Claude-3-Opus, and Gemini-1.5-Pro. Our code, data, and models are available at https://github.com/dvlab-research/Step-DPO.

For agents

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

get_harvested_code_for_paper("2406.18629")
get_code_for_paper("2406.18629")
have("2406.18629")

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

curl https://syntology.ai/api/ran/2406.18629.json

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

[![Syntology run record](https://syntology.ai/api/ran/2406.18629.svg)](https://syntology.ai/paper/2406.18629)

Connect an agent — have() is free.