SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2010.07543 · EMNLP Findings · 2020

Improving Constituency Parsing with Span Attention

Tong Zhang, Fei Xia, Yuanhe Tian, Yan Song ♠♥ †, Kong Shenzhen

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 (1 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; 0 no recorded run), measured 2026-11-22; 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/2010.07543.json · badge: /api/ran/2010.07543.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
cuhksz-nlp/SAPar marked official in Papers with Code
Papers with Code archive
1 ran · 0 no recorded run
FunctionStatusWhere it lives
FindNgrams Ran
ran on a synthesized input · constructed an object; no result was computed
cuhksz-nlp/SAPar/attutil.py
code served (permissive licence) · get_code("4147856ea4d1a0a6")
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 an instance of its own class, FindNgrams: the run constructed the object and did not call it, so no result was computed. 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 re
import math
from collections import defaultdict

class FindNgrams:

    def __init__(self, min_count=0, min_pmi=0, language='en'):
        self.min_count = min_count
        self.min_pmi = min_pmi
        self.words = defaultdict(int)
        self.ngrams, self.pairs = (defaultdict(int), defaultdict(int))
        self.total = 0.0
        self.language = language

    def text_filter(self, sentence):
        cleaned_text = []
        index = 0
        for i, w in enumerate(sentence):
            if re.match(u'[^\u0600-ۿݐ-ݿ一-龥0-9a-zA-Z]+', w):
                if i > index:
                    cleaned_text.append([w.lower() for w in sentence[index:i]])
                index = 1 + i
        if index < len(sentence):
            cleaned_text.append([w.lower() for w in sentence[index:]])
        return cleaned_text

    def count_ngram(self, texts, n):
        self.ngrams = defaultdict(int)
        for sentence in texts:
            for sub_sentence in self.text_filter(sentence):
                for i in range(n):
                    n_len = i + 1
                    for j in range(len(sub_sentence) - i):
                        ngram = tuple([w for w in sub_sentence[j:j + n_len]])
                        self.ngrams[ngram] += 1
        self.ngrams = {i: j for i, j in self.ngrams.items() if j > self.min_count}

    def find_ngrams_pmi(self, texts, n, freq_threshold):
        for sentence in texts:
            for sub_sentence in self.text_filter(sentence):
                self.words[sub_sentence[0]] += 1
                for i in range(len(sub_sentence) - 1):
                    self.words[sub_sentence[i + 1]] += 1
                    self.pairs[sub_sentence[i], sub_sentence[i + 1]] += 1
                    self.total += 1
        self.words = {i: j for i, j in self.words.items() if j > self.min_count}
        self.pairs = {i: j for i, j in self.pairs.items() if j > self.min_count}
        min_mi = math.inf
        max_mi = -math.inf
        self.strong_segments = set()
        for i, j in self.pairs.items():
            if i[0] in self.words and i[1] in self.words:
                mi = math.log(self.total * j / (self.words[i[0]] * self.words[i[1]]))
                if mi > max_mi:
                    max_mi = mi
                if mi < min_mi:
                    min_mi = mi
                if mi >= self.min_pmi:
                    self.strong_segments.add(i)
        self.ngrams = defaultdict(int)
        for sentence in texts:
            for sub_sentence in self.text_filter(sentence):
                s = [sub_sentence[0]]
                for i in range(len(sub_sentence) - 1):
                    if (sub_sentence[i], sub_sentence[i + 1]) in self.strong_segments:
                        s.append(sub_sentence[i + 1])
                    else:
                        self.ngrams[tuple(s)] += 1
                        s = [sub_sentence[i + 1]]
        self.ngrams = {i: j for i, j in self.ngrams.items() if j > self.min_count and len(i) <= n}
        self.renew_ngram_by_freq(texts, freq_threshold, n)

    def renew_ngram_by_freq(self, all_sentences, min_feq, ngram_len=10):
        new_ngram2count = {}
        new_all_sentences = []
        for sentence in all_sentences:
            for sen in self.text_filter(sentence):
                for i in range(len(sen)):
                    for n in range(1, ngram_len + 1):
                        if i + n > len(sentence):
                            break
                        n_gram = tuple(sentence[i:i + n])
                        if n_gram not in self.ngrams:
                            continue
                        if n_gram not in new_ngram2count:
                            new_ngram2count[n_gram] = 1
                        else:
                            new_ngram2count[n_gram] += 1
        self.ngrams = {gram: c for gram, c in new_ngram2count.items() if c > min_feq}


ARGS = {
    'min_count': 0,
    'min_pmi': 0.0,
    'language': 'en'
}
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 FindNgrams — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a FindNgrams 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

Constituency parsing is a fundamental and important task for natural language understanding, where a good representation of contextual information can help this task. N-grams, which is a conventional type of feature for contextual information, have been demonstrated to be useful in many tasks, and thus could also be beneficial for constituency parsing if they are appropriately modeled. In this paper, we propose span attention for neural chartbased constituency parsing to leverage n-gram information. Considering that current chartbased parsers with Transformer-based encoder represent spans by subtraction of the hidden states at the span boundaries, which may cause information loss especially for long spans, we incorporate n-grams into span representations by weighting them according to their contributions to the parsing process. Moreover, we propose categorical span attention to further enhance the model by weighting ngrams within different length categories, and thus benefit long-sentence parsing. Experimental results on three widely used benchmark datasets demonstrate the effectiveness of our approach in parsing Arabic, Chinese, and English, where state-of-the-art performance is obtained by our approach on all of them. 1 † Corresponding author. 1 Our code and the best performing models are released at https://github.com/cuhksz-nlp/SAPar.

For agents

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

get_harvested_code_for_paper("2010.07543")
get_code_for_paper("2010.07543")
have("2010.07543")

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

curl https://syntology.ai/api/ran/2010.07543.json

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

[![Syntology run record](https://syntology.ai/api/ran/2010.07543.svg)](https://syntology.ai/paper/2010.07543)

Connect an agent — have() is free.