SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1808.08745 · 2018

Don't Give Me the Details, Just the Summary! Topic-Aware Convolutional Neural Networks for Extreme Summarization

Shay Cohen, Mirella Lapata, Shashi Narayan

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 4 have no contract verdict; 0 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/1808.08745.json · badge: /api/ran/1808.08745.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
shashiongithub/XSum canonical (classifier label)
classifier basis not recorded on the link
2 ran · 0 no recorded run
pltrdy/extoracle_summarization community (classifier label)
classifier basis not recorded on the link
2 ran · 0 no recorded run
FunctionStatusWhere it lives
get_training_stats Ran
ran on a synthesized input
shashiongithub/XSum/XSum-ConvS2S/singleprocess_train.py
code served (permissive licence) · get_code("d10052dcb1c9b468")
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 OrderedDict. 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 collections
import math

class Meter:
    """Simple meter that tracks average and count."""
    def __init__(self, avg=0.0, count=0):
        self.avg = avg
        self.count = count

class Trainer:
    """Minimal trainer class that implements the interface used by get_training_stats."""
    def __init__(self):
        self.meters = collections.OrderedDict()
        self.meters['train_loss'] = Meter(avg=0.5, count=10)
        self.meters['train_nll_loss'] = Meter(avg=0.45, count=10)
        self.meters['wps'] = Meter(avg=1250.5, count=10)
        self.meters['ups'] = Meter(avg=3.7, count=10)
        self.meters['wpb'] = Meter(avg=256.3, count=10)
        self.meters['bsz'] = Meter(avg=32.1, count=10)
        self.meters['gnorm'] = Meter(avg=0.123, count=10)
        self.meters['clip'] = Meter(avg=0.15, count=10)
        self.meters['oom'] = Meter(avg=0.0, count=10)
        self._num_updates = 42
        self._lr = 0.0005
    
    def get_meter(self, name):
        """Return the meter for the given name."""
        return self.meters[name]
    
    def get_num_updates(self):
        """Return the number of updates."""
        return self._num_updates
    
    def get_lr(self):
        """Return the learning rate."""
        return self._lr

trainer = Trainer()

ARGS = {
    'trainer': trainer
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type OrderedDict — a OrderedDict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a OrderedDict 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
get_valid_stats Ran
ran on a synthesized input
shashiongithub/XSum/XSum-ConvS2S/singleprocess_train.py
code served (permissive licence) · get_code("1f42a3abd8d7fae5")
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 OrderedDict. 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 collections
import math
import torch

class AverageMeter:
    """Simple meter to track average values"""
    def __init__(self):
        self.sum = 0.0
        self.count = 0
    
    @property
    def avg(self):
        if self.count == 0:
            return 0.0
        return self.sum / self.count
    
    def update(self, val, n=1):
        self.sum += val * n
        self.count += n

class MockTrainer:
    """Minimal trainer class with get_meter method"""
    def __init__(self):
        self.meters = {}
        
        # Initialize valid_loss meter
        self.meters['valid_loss'] = AverageMeter()
        self.meters['valid_loss'].update(2.5)
        
        # Initialize valid_nll_loss meter with count > 0
        self.meters['valid_nll_loss'] = AverageMeter()
        self.meters['valid_nll_loss'].update(2.3)
    
    def get_meter(self, name):
        return self.meters.get(name, AverageMeter())

torch.manual_seed(0)

trainer = MockTrainer()

ARGS = {
    'trainer': trainer
}
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 OrderedDict — a OrderedDict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a OrderedDict 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
process_example Ran
ran on a synthesized input
pltrdy/extoracle_summarization/extoracle/extoracle.py
code served (permissive licence) · get_code("55732ffb0494d417")
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 tuple. 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 numpy as np

def _clean_line(line):
    return line.strip()

def split_text(txt, trunc=None):
    """Split text into sentences/words

    Args:
        txt(str): text, as a single str
        trunc(int): if not None, stop splitting text after `trunc` words
                    and ignore sentence containing `trunc`-th word
                    (i.e. each sentence has len <= trunc)
    Returns:
        sentences(list): list of sentences (= list of lists of words)
    """
    n_words = 0
    sentence_break = ['.', '?', '!']
    sentences = []
    cur_sentence = []
    for word in txt.split():
        if word in sentence_break:
            if len(cur_sentence) != 0:
                cur_sentence.append(word)
                sentences.append(cur_sentence)
                cur_sentence = []
            else:
                pass
        else:
            n_words += 1
            cur_sentence.append(word)
        if trunc is not None and n_words > trunc:
            cur_sentence = []
            break
    if len(cur_sentence) != 0:
        sentences.append(cur_sentence)
    return sentences

def process_example(example):
    method, src_line, tgt_line, trunc_src, summary_length, length_oracle = example
    src_line = _clean_line(src_line)
    tgt_line = _clean_line(tgt_line)
    src_sentences = split_text(src_line, trunc=trunc_src)
    tgt_sentences = split_text(tgt_line)
    if length_oracle:
        summary_length = len(tgt_sentences)
    ids, sents = method(src_sentences, tgt_sentences, summary_length)
    return (ids, sents)

class MockMethod:
    def __call__(self, src_sentences, tgt_sentences, summary_length):
        ids = list(range(len(src_sentences)))
        sents = src_sentences[:summary_length]
        return (ids, sents)

mock_method = MockMethod()

ARGS = {
    "example": (
        mock_method,
        "  This is a source sentence.  ",
        "  This is a target sentence.  ",
        5,
        2,
        False
    )
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple — a tuple output has no array shape of its own; none is 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_text Ran
ran on a synthesized input
pltrdy/extoracle_summarization/extoracle/extoracle.py
code served (permissive licence) · get_code("8e675d0bebfa35fb")
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 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

# Define ARGS dict with valid arguments for split_text function
ARGS = {
    'txt': "Hello world. This is a test! How are you? I am fine.",
    'trunc': None
}
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 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, 2 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 2 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.

FunctionAlso present inEarliest-datedOthers named
get_training_stats 1 other paper
1 later
2005.00661 (2020-05)
On Faithfulness and Factuality in Abstractive Summarization
get_valid_stats 1 other paper
1 later
2005.00661 (2020-05)
On Faithfulness and Factuality in Abstractive Summarization

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

We introduce extreme summarization, a new single-document summarization task which does not favor extractive strategies and calls for an abstractive modeling approach. The idea is to create a short, one-sentence news summary answering the question "What is the article about?". We collect a real-world, large scale dataset for this task by harvesting online articles from the British Broadcasting Corporation (BBC). We propose a novel abstractive model which is conditioned on the article's topics and based entirely on convolutional neural networks. We demonstrate experimentally that this architecture captures longrange dependencies in a document and recognizes pertinent content, outperforming an oracle extractive system and state-of-the-art abstractive approaches when evaluated automatically and by humans. 1

For agents

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

get_harvested_code_for_paper("1808.08745")
get_code_for_paper("1808.08745")
have("1808.08745")

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

curl https://syntology.ai/api/ran/1808.08745.json

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

[![Syntology run record](https://syntology.ai/api/ran/1808.08745.svg)](https://syntology.ai/paper/1808.08745)

Connect an agent — have() is free.