SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2602.07038 · 2026

UNIKIE-BENCH: Benchmarking Large Multimodal Models for Key Information Extraction in Visual Documents

Qian Zhang, Maosong Sun, Junyang Lin, Zhenghao Liu, Yu Gu, Zulong Chen, Zhipeng Xu, Zhibo Yang, Large Model, Yifan Ji

arXiv · PDF · Open in the Atlas

Code that ran

3 of 3 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; 3 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/2602.07038.json · badge: /api/ran/2602.07038.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
NEUIR/UNIKIE-BENCH community
Papers with Code archive
3 ran · 0 no recorded run
FunctionStatusWhere it lives
cal_f1_all Ran
ran; our contract draft was wrong, not the code
NEUIR/UNIKIE-BENCH/src/kie_evaluator.py
code served (permissive licence) · get_code("54cf98bf3989cfc1")
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, 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.

# Setup code for cal_f1_all function

def setup_test_data():
    """Create minimal test data for cal_f1_all"""
    
    # Create predictions dict with nested structure
    preds = {
        "document_1": {
            "menu": [
                {"name": ["cake"], "count": ["2"]},
                {"name": ["juice"], "count": ["1"]},
            ]
        },
        "document_2": {
            "invoice": [
                {"total": ["100"], "date": ["2024-01-15"]},
            ]
        }
    }
    
    # Create answers dict (ground truth) with nested structure
    answers = {
        "document_1": {
            "menu": [
                {"name": ["cake"], "count": ["2"]},
                {"name": ["juice"], "count": ["1"]},
            ]
        },
        "document_2": {
            "invoice": [
                {"total": ["100"], "date": ["2024-01-15"]},
            ]
        }
    }
    
    return preds, answers

preds, answers = setup_test_data()

ARGS = {
    "preds": preds,
    "answers": answers,
}
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 tuple, ndim 1, dtype family float — a tuple 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
flatten Ran
ran; our contract draft was wrong, not the code
NEUIR/UNIKIE-BENCH/src/kie_evaluator.py
code served (permissive licence) · get_code("75bdbe1b0b5829c3")
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
import numpy as np

# Construct a valid nested dictionary for the flatten function
data = {
    "menu": [
        {"name": ["cake"], "count": ["2"]},
        {"name": ["juice"], "count": ["1"]},
    ]
}

ARGS = {
    "data": data
}
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 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
normalize_dict Ran
ran; our contract draft was wrong, not the code
NEUIR/UNIKIE-BENCH/src/kie_evaluator.py
code served (permissive licence) · get_code("031a1069599e8260")
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 dict. 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

torch.manual_seed(0)

# Test case 1: Complex nested dictionary with various types
data_dict = {
    "model": "resnet",
    "a": 42,
    "learning_rate": 0.001,
    "layers": [1, 2, 3],
    "config": {
        "z": "value",
        "b": [4.5, 6.7],
        "nested": {
            "x": "test"
        }
    },
    "empty_val": ""
}

ARGS = {
    "data": data_dict
}
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 dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict 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 3 samples lifted from this paper's linked repositories, 3 carry a function identity and 0 do not. Of the 3 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 3 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

Key Information Extraction (KIE) from realworld documents remains challenging due to substantial variations in layout structures, visual quality, and task-specific information requirements. Recent Large Multimodal Models (LMMs) have shown promising potential for performing end-to-end KIE directly from document images. To enable a comprehensive and systematic evaluation across realistic and diverse application scenarios, we introduce UNIKIE-BENCH, a unified benchmark designed to rigorously evaluate the KIE capabilities of LMMs. UNIKIE-BENCH consists of two complementary tracks: a constrainedcategory KIE track with scenario-predefined schemas that reflect practical application needs, and an open-category KIE track that extracts any key information that is explicitly present in the document. Experiments on 15 stateof-the-art LMMs reveal substantial performance degradation under diverse schema definitions, long-tail key fields, and complex layouts, along with pronounced performance disparities across different document types and scenarios. These findings underscore persistent challenges in grounding accuracy and layout-aware reasoning for LMM-based KIE. All codes and datasets are available at https: //github.com/NEUIR/UNIKIE-BENCH. * indicates equal contribution. † indicates corresponding author.

For agents

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

get_harvested_code_for_paper("2602.07038")
get_code_for_paper("2602.07038")
have("2602.07038")

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

curl https://syntology.ai/api/ran/2602.07038.json

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

[![Syntology run record](https://syntology.ai/api/ran/2602.07038.svg)](https://syntology.ai/paper/2602.07038)

Connect an agent — have() is free.