SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2406.07971 · 2024

It Takes Two: On the Seamlessness between Reward and Policy Model in RLHF

arXiv · PDF · Open in the Atlas

Code that ran

12 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; 3 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 9 failures of our instrument; 0 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/2406.07971.json · badge: /api/ran/2406.07971.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
taiminglu/seamless canonical (classifier label)
classifier basis: census
12 ran · 0 no recorded run
FunctionStatusWhere it lives
calculate_log_probability Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
taiminglu/seamless/code/SEAM/policy/get_prob.py
pointer only (licence: NONE) · get_code("e1333323e0d954bc")
How it ran

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

collator Ran
ran; our contract draft was wrong, not the code
taiminglu/seamless/code/train/rl.py
pointer only (licence: NONE) · get_code("fddcce152bc7d3bd")
How it ran

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

combine_and_pad_tokens Ran
ran, honoured the contract we drafted
taiminglu/seamless/code/SEAM/policy/get_prob.py
pointer only (licence: NONE) · get_code("f8f1cb4c32403020")
How it ran

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

encode_and_search Ran
ran, honoured the contract we drafted
taiminglu/seamless/code/retrival/contrast/retrival.py
pointer only (licence: NONE) · get_code("31218514d492527d")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, taiminglu/seamless 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_completion Ran
ran; our contract draft was wrong, not the code
taiminglu/seamless/code/retrival/gpt/retrival.py
pointer only (licence: NONE) · get_code("fb97e66f814910e9")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, taiminglu/seamless 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_prompt Ran
ran; our contract draft was wrong, not the code
taiminglu/seamless/code/retrival/gpt/retrival.py
pointer only (licence: NONE) · get_code("845ccc17a73fe529")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, taiminglu/seamless 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_results Ran
ran; our contract draft was wrong, not the code
taiminglu/seamless/code/retrival/adv/retrival.py
pointer only (licence: NONE) · get_code("48fc27d8ef2d0595")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, taiminglu/seamless 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_reward Ran
ran, honoured the contract we drafted
taiminglu/seamless/code/SEAM/reward/get_reward.py
pointer only (licence: NONE) · get_code("646f36a115dc20f4")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, taiminglu/seamless 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_text Ran
ran; our contract draft was wrong, not the code
taiminglu/seamless/code/SEAM/reward/get_reward.py
pointer only (licence: NONE) · get_code("55737ec700776650")
How it ran

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

load_data Ran
ran; our contract draft was wrong, not the code
taiminglu/seamless/code/retrival/contrast/retrival.py
pointer only (licence: NONE) · get_code("43b18b0788bb9a75")
How it ran

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

load_data Ran
ran; our contract draft was wrong, not the code
taiminglu/seamless/code/retrival/gpt/retrival.py
pointer only (licence: NONE) · get_code("14d77a6c7cf50243")
How it ran

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

prepare_sample_text Ran
ran; our contract draft was wrong, not the code
taiminglu/seamless/code/train/sft.py
pointer only (licence: NONE) · get_code("148dc2c38b7cd922")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, taiminglu/seamless 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 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, 2 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 2 later) and 10 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
prepare_sample_text 3 other papers
2 dated earlier, 1 later
2309.16155 (2023-09)
The Trickle-down Impact of Reward (In-)consistency on RLHF
2405.01511 (2024-05)
D2PO: Discriminator-Guided DPO with Response Evaluation Models
2407.12043 (2024-07)
The Art of Saying No: Contextual Noncompliance in Language Models
collator 2 other papers
1 dated earlier, 1 later
2405.06639 (2024-05)
Value Augmented Sampling for Language Model Alignment and Personalization
2407.03856 (2024-07)
Q-Adapter: Customizing Pre-trained LLMs to New Preferences with Forgetting Mitigation

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

Reinforcement Learning from Human Feedback (RLHF) involves training policy models (PMs) and reward models (RMs) to align language models with human preferences. Instead of focusing solely on PMs and RMs independently, we propose to examine their interactions during fine-tuning, introducing the concept of seamlessness. Our study starts with observing the saturation phenomenon, where continual improvements in RM and PM do not translate into RLHF progress. Our analysis shows that RMs fail to assign proper scores to PM responses, resulting in a 35% mismatch rate with human preferences, highlighting a significant discrepancy between PM and RM. To measure seamlessness between PM and RM without human effort, we propose an automatic metric, SEAM. SEAM quantifies the discrepancies between PM and RM judgments induced by data samples. We validate the effectiveness of SEAM in data selection and model augmentation. Our experiments demonstrate that (1) using SEAM-filtered data for RL training improves RLHF performance by 4.5%, and (2) SEAM-guided model augmentation results in a 4% performance improvement over standard augmentation methods.

For agents

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

get_harvested_code_for_paper("2406.07971")
get_code_for_paper("2406.07971")
have("2406.07971")

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

curl https://syntology.ai/api/ran/2406.07971.json

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

[![Syntology run record](https://syntology.ai/api/ran/2406.07971.svg)](https://syntology.ai/paper/2406.07971)

Connect an agent — have() is free.