SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2410.11287 · 2024

Process Reward Model with Q-Value Rankings

arXiv · PDF · Open in the Atlas

Code that ran

4 of 4 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 0 have no contract verdict; 4 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/2410.11287.json · badge: /api/ran/2410.11287.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
WindyLee0822/Process_Q_Model canonical (classifier label)
classifier basis: classifier_2025
4 ran · 0 no recorded run
FunctionStatusWhere it lives
best_of_n Ran
ran; our contract draft was wrong, not the code
WindyLee0822/Process_Q_Model/bon_eval_hf.py
code served (permissive licence) · get_code("f4487de774c49640")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

# The function expects splitted_completions, which is an iterable of groups
# where each group contains completions (dicts with 'reward' key)
# The function sorts by reward and selects the best one from each group

splitted_completions = [
    [
        {'reward': 0.5, 'text': 'completion 1'},
        {'reward': 0.8, 'text': 'completion 2'},
        {'reward': 0.3, 'text': 'completion 3'},
    ],
    [
        {'reward': 0.9, 'text': 'completion 4'},
        {'reward': 0.6, 'text': 'completion 5'},
    ],
    [
        {'reward': 0.7, 'text': 'completion 6'},
        {'reward': 0.4, 'text': 'completion 7'},
        {'reward': 0.2, 'text': 'completion 8'},
        {'reward': 0.95, 'text': 'completion 9'},
    ],
]

ARGS = {
    'splitted_completions': splitted_completions
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
instruction_format Ran
ran; our contract draft was wrong, not the code
WindyLee0822/Process_Q_Model/train_main.py
code served (permissive licence) · get_code("47c7df00c43e4f1a")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned str. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import torch.nn as nn

torch.manual_seed(0)

# The function instruction_format takes a single parameter 's' which is a string
# It simply wraps the string with [INST] and [/INST] markers
# No complex objects needed - just a string

ARGS = {
    "s": "What is the capital of France?"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
instruction_format Ran
ran; our contract draft was wrong, not the code
WindyLee0822/Process_Q_Model/bon_eval_hf.py
code served (permissive licence) · get_code("262c7067e52c3a8c")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned str. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

def instruction_format(s):
    return f"Below is an instruction that describes a task.\nWrite a response that appropriately completes the request.\n\n### Instruction:\n{s}\n\n### Response: Let's think step by step"

ARGS = {
    "s": "What is the capital of France?"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
split_query Ran
ran; our contract draft was wrong, not the code
WindyLee0822/Process_Q_Model/bon_eval_hf.py
code served (permissive licence) · get_code("29a9b4056106f7db")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1) and returned list, ndim 2, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np

torch.manual_seed(0)

# Construct completions list
# Each completion is a dict with 'idx' and 'logprobs' keys
completions = [
    {'idx': 0, 'logprobs': 0.9},
    {'idx': 0, 'logprobs': 0.7},
    {'idx': 0, 'logprobs': 0.5},
    {'idx': 0, 'logprobs': 0.3},
    {'idx': 1, 'logprobs': 0.8},
    {'idx': 1, 'logprobs': 0.6},
    {'idx': 1, 'logprobs': 0.4},
    {'idx': 1, 'logprobs': 0.2},
    {'idx': 2, 'logprobs': 0.95},
    {'idx': 2, 'logprobs': 0.75},
    {'idx': 2, 'logprobs': 0.55},
    {'idx': 2, 'logprobs': 0.25},
    {'idx': 3, 'logprobs': 0.85},
    {'idx': 3, 'logprobs': 0.65},
    {'idx': 3, 'logprobs': 0.45},
    {'idx': 3, 'logprobs': 0.15},
]

n = 2
N = 16

ARGS = {
    'completions': completions,
    'n': n,
    'N': N,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 2, dtype family float — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block

Also present in other papers' code

Of 4 samples lifted from this paper's linked repositories, 4 carry a function identity and 0 do not. Of the 4 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 4 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 2026-11-22.

Repositories linked to this paper

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

Abstract

Process Reward Modeling (PRM) is critical for complex reasoning and decision-making tasks where the accuracy of intermediate steps significantly influences the overall outcome. Existing PRM approaches, primarily framed as classification problems, employ cross-entropy loss to independently evaluate each step's correctness. This method can lead to suboptimal reward distribution and does not adequately address the interdependencies among steps. To address these limitations, we introduce the Process Q-value Model (PQM), a novel framework that redefines PRM in the context of a Markov Decision Process. PQM optimizes Q-value rankings based on a novel comparative loss function, enhancing the model's ability to capture the intricate dynamics among sequential decisions. This approach provides a more granular and theoretically grounded methodology for process rewards. Our extensive empirical evaluations across various sampling policies, language model backbones, and multi-step reasoning benchmarks show that PQM outperforms classification-based PRMs. The effectiveness of the comparative loss function is highlighted in our comprehensive ablation studies, confirming PQM's practical efficacy and theoretical advantage.

For agents

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

get_harvested_code_for_paper("2410.11287")
get_code_for_paper("2410.11287")
have("2410.11287")

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

curl https://syntology.ai/api/ran/2410.11287.json

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

[![Syntology run record](https://syntology.ai/api/ran/2410.11287.svg)](https://syntology.ai/paper/2410.11287)

Connect an agent — have() is free.