SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2311.05565 · NeurIPS · 2023

High-Performance Transformers for Table Structure Recognition Need Early Convolutions

Duen Chau, Shengyun Peng, Seongmin Lee, Xiaojing Wang, Rajarajeswari Balasubramaniyan

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 0 have no contract verdict; 1 failure 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/2311.05565.json · badge: /api/ran/2311.05565.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
poloclub/tsr-convstem canonical (classifier label)
classifier basis: census
1 ran · 1 no recorded run
FunctionStatusWhere it lives
parallel_process Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
poloclub/tsr-convstem/src/utils/teds.py
code served (permissive licence) · get_code("de9e6c4743b42946")
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 numpy as np
from typing import List, Callable, Any

# Define a simple function that works with single items
def simple_square(x: int) -> int:
    """Simple function that squares a number"""
    return x ** 2

# Define a function that works with kwargs
def kwargs_multiply(a: int, b: int) -> int:
    """Function that multiplies two kwargs"""
    return a * b

# Create test array with integers
test_array = [1, 2, 3, 4, 5]

# Create test array with dictionaries for kwargs usage
test_kwargs_array = [
    {"a": 1, "b": 2},
    {"a": 3, "b": 4},
    {"a": 5, "b": 6},
]

# Define ARGS dictionary with all parameters
ARGS = {
    "array": test_array,
    "function": simple_square,
    "n_jobs": 2,
    "use_kwargs": False,
    "front_num": 1,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch, tqdm — 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
combine_html_table_structure_text Not yet run
unverified (harvested, no recorded run)
poloclub/tsr-convstem/src/utils/utils.py
code served (permissive licence) · get_code("eaa8420ba2c27ae6")
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, 1 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 1 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.

FunctionAlso present inEarliest-datedOthers named
parallel_process 3 other papers
3 later
2405.11862 (2024-05)
SEMv3: A Fast and Robust Approach to Table Separation Line Detection
2501.00321 (2025-01)
OCRBench v2: An Improved Benchmark for Evaluating Large Multimodal Models on Visual Text Localization and Reasoning
2607.02956 (2026-07)
MORE: A Multilingual Document Parsing Benchmark and Evaluation

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

Table structure recognition (TSR) aims to convert tabular images into a machinereadable format, where a visual encoder extracts image features and a textual decoder generates table-representing tokens. Existing approaches use classic convolutional neural network (CNN) backbones for the visual encoder and transformers for the textual decoder. However, this hybrid CNN-Transformer architecture introduces a complex visual encoder that accounts for nearly half of the total model parameters, markedly reduces both training and inference speed, and hinders the potential for self-supervised learning in TSR. In this work, we design a lightweight visual encoder for TSR without sacrificing expressive power. We discover that a convolutional stem can match classic CNN backbone performance, with a much simpler model. The convolutional stem strikes an optimal balance between two crucial factors for high-performance TSR: a higher receptive field (RF) ratio and a longer sequence length. This allows it to "see" an appropriate portion of the table and "store" the complex table structure within sufficient context length for the subsequent transformer. We conducted reproducible ablation studies and open-sourced our code at https://github.com/poloclub/tsr-convstem to enhance transparency, inspire innovations, and facilitate fair comparisons in our domain as tables are a promising modality for representation learning.

For agents

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

get_harvested_code_for_paper("2311.05565")
get_code_for_paper("2311.05565")
have("2311.05565")

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

curl https://syntology.ai/api/ran/2311.05565.json

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

[![Syntology run record](https://syntology.ai/api/ran/2311.05565.svg)](https://syntology.ai/paper/2311.05565)

Connect an agent — have() is free.