SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2211.06196 · EMNLP · 2022

Improving Factual Consistency in Summarization with Compression-Based Post-Editing

Chien-Sheng Wu, Caiming Xiong, Jesse Vig, Kumar Choubey, Alexander Fabbri Prafulla

arXiv · PDF · Open in the Atlas

Code that ran

1 of 2 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 1 has no contract verdict; 0 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/2211.06196.json · badge: /api/ran/2211.06196.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
salesforce/CompEdit canonical (classifier label)
classifier basis not recorded on the link
1 ran · 1 no recorded run
FunctionStatusWhere it lives
get_gen Ran
ran on a synthesized input
salesforce/CompEdit/generate.py
code served (permissive licence) · get_code("7d8482b7338b18ee")
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 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
import torch.nn as nn

torch.manual_seed(0)

class MockTokenizer:
    def __call__(self, text, max_length=None, truncation=False, padding=None, return_tensors=None):
        batch_size = len(text) if isinstance(text, list) else 1
        return {
            'input_ids': torch.randint(0, 100, (batch_size, max_length or 10)),
            'attention_mask': torch.ones(batch_size, max_length or 10)
        }
    
    def decode(self, token_ids, skip_special_tokens=False, clean_up_tokenization_spaces=False):
        return "This is a decoded summary with <n> special token"

class MockGenerativeModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.linear = nn.Linear(4, 4)
    
    def generate(self, input_ids, attention_mask, length_penalty=None, num_beams=None, max_length=None, min_length=None):
        batch_size = input_ids.shape[0]
        return torch.randint(0, 100, (batch_size, max_length or 20))

inputs = ["This is a test document for summarization.", "Another test input for the model."]
model = MockGenerativeModel()
tokenizer = MockTokenizer()

ARGS = {
    'inputs': inputs,
    'model': model,
    'tokenizer': tokenizer,
    'args_dict': {
        'max_enc_len': 10,
        'max_gen_len': 20,
        'min_gen_len': 5,
        'length_penalty': 2.0,
        'num_beams': 4,
        'device': 'cpu'
    }
}
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 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
process_file Not yet run
unverified (harvested, no recorded run)
salesforce/CompEdit/entity_score.py
code served (permissive licence) · get_code("ad8f0274bbfb39b6")
How it ran

no invocation recorded.

Also present in other papers' code

Of 2 samples lifted from this paper's linked repositories, 2 carry a function identity and 0 do not. Of the 2 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 2 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

State-of-the-art summarization models still struggle to be factually consistent with the input text. A model-agnostic way to address this problem is post-editing the generated summaries. However, existing approaches typically fail to remove entity errors if a suitable input entity replacement is not available or may insert erroneous content. In our work, we focus on removing extrinsic entity errors, or entities not in the source, to improve consistency while retaining the summary's essential information and form. We propose to use sentence-compression data to train the post-editing model to take a summary with extrinsic entity errors marked with special tokens and output a compressed, well-formed summary with those errors removed. We show that this model improves factual consistency while maintaining ROUGE, improving entity precision by up to 30% on XSum, and that this model can be applied on top of another post-editor, improving entity precision by up to a total of 38%. We perform an extensive comparison of post-editing approaches that demonstrate trade-offs between factual consistency, informativeness, and grammaticality, and we analyze settings where posteditors show the largest improvements. Source: The oil price collapse sent global markets reeling throughout 2015. ... Brent crude oil was up 3% at $37.60 per barrel for the day but down 35% over the year. System Summary Original Wall Street markets closed lower on the last trading day of 2015 as oil prices languished at $28 a barrel for much of the year. Entity Swap No Change Rewrite Wall Street markets closed lower on the last trading day of 2015 as oil prices languished at $37.60 per barrel ... but remain down 35% over the year. Delete Wall Street markets closed lower on 2015 as oil prices languished at a for much of the year. Compress (ours) Wall Street markets closed lower as oil prices languished for much of the year.

For agents

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

get_harvested_code_for_paper("2211.06196")
get_code_for_paper("2211.06196")
have("2211.06196")

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

curl https://syntology.ai/api/ran/2211.06196.json

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

[![Syntology run record](https://syntology.ai/api/ran/2211.06196.svg)](https://syntology.ai/paper/2211.06196)

Connect an agent — have() is free.