SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2407.02397 · 2024

Learning to Refine with Fine-Grained Natural Language Feedback

arXiv · PDF · Open in the Atlas

Code that ran

1 of 1 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; 1 failure 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/2407.02397.json · badge: /api/ran/2407.02397.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
manyawadhwa/dcr canonical (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
FunctionStatusWhere it lives
resize_embedding Ran
ran; our contract draft was wrong, not the code
manyawadhwa/dcr/run_end_to_end_refinement/dcr.py
code served (permissive licence) · get_code("ee48cefe00da7f08")
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, through the batch API) and returned StandInModelWrapper. 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)

class StandInTokenizer:
    def __init__(self):
        self.vocab_size = 100
        self.padding_side = 'left'
    
    def add_special_tokens(self, special_tokens_dict):
        if 'pad_token' in special_tokens_dict:
            self.vocab_size += 1
    
    def __len__(self):
        return self.vocab_size

class StandInModel(nn.Module):
    def __init__(self, vocab_size=101):
        super().__init__()
        self.embedding = nn.Embedding(vocab_size, 4)
    
    def resize_token_embeddings(self, new_vocab_size):
        old_embedding = self.embedding
        self.embedding = nn.Embedding(new_vocab_size, 4)
        return self.embedding

class StandInModelWrapper(nn.Module):
    def __init__(self):
        super().__init__()
        self.tokenizer = StandInTokenizer()
        self.model = StandInModel(vocab_size=len(self.tokenizer))
    
    def forward(self, x):
        return self.model.embedding(x)

model = StandInModelWrapper()

ARGS = {
    'model': model
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type StandInModelWrapper — a StandInModelWrapper output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a StandInModelWrapper 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

Also present in other papers' code

Of 1 samples lifted from this paper's linked repositories, 1 carry a function identity and 0 do not. Of the 1 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 1 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

Recent work has explored the capability of large language models (LLMs) to identify and correct errors in LLM-generated responses. These refinement approaches frequently evaluate what sizes of models are able to do refinement for what problems, but less attention is paid to what effective feedback for refinement looks like. In this work, we propose looking at refinement with feedback as a composition of three distinct LLM competencies: (1) detection of bad generations; (2) fine-grained natural language critique generation; (3) refining with fine-grained feedback. The first step can be implemented with a high-performing discriminative model and steps 2 and 3 can be implemented either via prompted or fine-tuned LLMs. A key property of the proposed Detect, Critique, Refine ("DCR") method is that the step 2 critique model can give fine-grained feedback about errors, made possible by offloading the discrimination to a separate model in step 1. We show that models of different capabilities benefit from refining with DCR on the task of improving factual consistency of document grounded summaries. Overall, DCR consistently outperforms existing end-to-end refinement approaches and current trained models not fine-tuned for factuality critiquing.

For agents

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

get_harvested_code_for_paper("2407.02397")
get_code_for_paper("2407.02397")
have("2407.02397")

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

curl https://syntology.ai/api/ran/2407.02397.json

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

[![Syntology run record](https://syntology.ai/api/ran/2407.02397.svg)](https://syntology.ai/paper/2407.02397)

Connect an agent — have() is free.