SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2603.08035 · 2026

CDRRM: Contrast-Driven Rubric Generation for Reliable and Interpretable Reward Modeling

Jiahao Li, Zhendong Mao, Dengcan Liu, Fengkai Yang

arXiv · PDF · Open in the Atlas

Code that ran

8 of 13 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, 0 violated it and 3 have no contract verdict; 5 failures of our instrument; 5 no recorded run), measured 2026-11-21; 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/2603.08035.json · badge: /api/ran/2603.08035.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
ldcan/CDRRM community
Papers with Code archive
8 ran · 5 no recorded run
FunctionStatusWhere it lives
DiagnosisResult Ran
ran on a synthesized input
ldcan/CDRRM/rubric_synthesis/rubric_generation.py
pointer only (licence: NONE) · get_code("77a699e6a17f2684")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ldcan/CDRRM 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.

RubricPairResult Ran
ran on a synthesized input
ldcan/CDRRM/rubric_synthesis/rubric_generation.py
pointer only (licence: NONE) · get_code("2107ef3bb57fc559")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ldcan/CDRRM 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.

_redact_label_fields Ran
ran; our contract draft was wrong, not the code
ldcan/CDRRM/rubric_synthesis/rubric_generation.py
pointer only (licence: NONE) · get_code("4e67311afdef640f")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ldcan/CDRRM 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_json_from_text Ran
ran; our contract draft was wrong, not the code
ldcan/CDRRM/rubric_synthesis/rubric_generation.py
pointer only (licence: NONE) · get_code("3658e51d3c3d4553")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ldcan/CDRRM 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_async_client Ran
ran on a synthesized input
ldcan/CDRRM/rubric_synthesis/rubric_generation.py
pointer only (licence: NONE) · get_code("6233c5ffe4a95226")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ldcan/CDRRM 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.

iter_jsonl Ran
ran; our contract draft was wrong, not the code
ldcan/CDRRM/rubric_synthesis/rubric_generation.py
pointer only (licence: NONE) · get_code("6db26f65525222c4")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ldcan/CDRRM 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.

llm_call Ran
ran; our contract draft was wrong, not the code
ldcan/CDRRM/rubric_synthesis/rubric_generation.py
pointer only (licence: NONE) · get_code("ef6061f563084d20")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ldcan/CDRRM 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_pair_data Ran
ran; our contract draft was wrong, not the code
ldcan/CDRRM/rubric_synthesis/rubric_generation.py
pointer only (licence: NONE) · get_code("c5bd7a3118ab0222")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ldcan/CDRRM 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.

_ensure_parent Not yet run
unverified (harvested, no recorded run)
ldcan/CDRRM/rubric_synthesis/rubric_generation.py
pointer only (licence: NONE) · get_code("1a1b7e4fdf687b67")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ldcan/CDRRM 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.

_load_diagnosis_index Not yet run
unverified (harvested, no recorded run)
ldcan/CDRRM/rubric_synthesis/rubric_generation.py
pointer only (licence: NONE) · get_code("b93615142c9738bb")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ldcan/CDRRM 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.

append_jsonl Not yet run
unverified (harvested, no recorded run)
ldcan/CDRRM/rubric_synthesis/rubric_generation.py
pointer only (licence: NONE) · get_code("4b5d4b04ddc2753c")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ldcan/CDRRM 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.

generate_rubric_async Not yet run
unverified (harvested, no recorded run)
ldcan/CDRRM/rubric_synthesis/rubric_generation.py
pointer only (licence: NONE) · get_code("9e9adff62748cb20")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ldcan/CDRRM 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.

generate_rubric_batch Not yet run
unverified (harvested, no recorded run)
ldcan/CDRRM/rubric_synthesis/rubric_generation.py
pointer only (licence: NONE) · get_code("7f4ed53e7f34a1cd")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ldcan/CDRRM 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 13 samples lifted from this paper's linked repositories, 13 carry a function identity and 0 do not. Of the 13 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 13 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.

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 2026-11-21.

Repositories linked to this paper

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

Abstract

Reward modeling is essential for aligning Large Language Models(LLMs) with human preferences, yet conventional reward models suffer from poor interpretability and heavy reliance on costly expert annotations. While recent rubricbased approaches enhance evaluation transparency, they lack systematic quality control, yielding noisy and redundant criteria, failing to mitigate persistent biases (e.g., verbosity, position) in LLM evaluators, and creating a scalabilityreliability trade-off. To address these limitations, we propose CDRRM (Contrast-Driven Rubric Reward Model), a framework built on a novel Contrast-then-Synthesis paradigm for high-quality rubric generation and guided preference judgment. CDRRM first conducts multi-dimensional contrastive profiling on preference pairs to identify causal discriminative factors, then synthesizes these insights into compact, context-aware rubrics to guide preference judgments. Extensive experiments on three authoritative benchmarks (RewardBench, RMBench, RMB) demonstrate that CDRRM achieves state-of-the-art performance across diverse domains and effectively mitigates aforementioned evaluation biases. Notably, our approach delivers exceptional data efficiency: training the rubric generator on only 3k high-quality samples empowers a frozen pre-trained judge model to outperform fully fine-tuned baselines. This work offers a scalable, interpretable, and data-efficient path for reward modeling.

For agents

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

get_harvested_code_for_paper("2603.08035")
get_code_for_paper("2603.08035")
have("2603.08035")

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

curl https://syntology.ai/api/ran/2603.08035.json

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

[![Syntology run record](https://syntology.ai/api/ran/2603.08035.svg)](https://syntology.ai/paper/2603.08035)

Connect an agent — have() is free.