SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2406.14852 · NeurIPS · 2024

Is A Picture Worth A Thousand Words? Delving Into Spatial Reasoning for Vision Language Models

Yixuan Li, Xin Wang, Yifei Ming, Vibhav Vineet, - Madison, Zhenmei Shi, Jiayu Wang, Neel Joshi

arXiv · PDF · Open in the Atlas

Code that ran

12 of 13 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; 1 honoured the contract we drafted, 1 violated it and 0 have no contract verdict; 10 failures of our instrument; 1 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/2406.14852.json · badge: /api/ran/2406.14852.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
BAAI-DCAI/Bunny canonical (classifier label)
classifier basis: census
6 ran · 1 no recorded run
jiayuww/SpatialEval canonical (classifier label)
classifier basis: classifier_2025
6 ran · 0 no recorded run
FunctionStatusWhere it lives
construct_prompt Ran
ran; our contract draft was wrong, not the code
BAAI-DCAI/Bunny/bunny/eval/model_vqa_cmmmu.py
code served (permissive licence) · get_code("2fd1673b298c086c")
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

torch.manual_seed(0)

# Define sample dict for a multiple choice question
sample_multi_choice = {
    'question': 'What is the capital of France?',
    'option1': 'Paris',
    'option2': 'London',
    'option3': 'Berlin',
    'option4': 'Madrid',
    'type': '选择',
    'answer': 'A',
}

# Define config dict with required format strings and instructions
config = {
    'multi_choice_example_format': 'Question: {}\nOptions:\n{}',
    'T/F_example_format': 'True/False Question: {}\n{}',
    'short_ans_example_format': 'Short Answer: {}',
    'task_instructions': [
        'Please answer the following multiple choice question.',
        'Please answer the following true/false question.',
        'Please provide a short answer to the following question.'
    ]
}

ARGS = {
    'sample': sample_multi_choice,
    'config': config
}
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 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
extract_before_is Ran
ran; our contract draft was wrong, not the code
jiayuww/SpatialEval/evals/evaluation.py
pointer only (licence: NONE) · get_code("e7d1140ae1160f39")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, jiayuww/SpatialEval when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

extract_model_name Ran
ran; our contract draft was wrong, not the code
jiayuww/SpatialEval/evals/evaluation.py
pointer only (licence: NONE) · get_code("d7410f697a9e72fb")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, jiayuww/SpatialEval when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

format_bunny_tqa_prompt_hf Ran
ran; our contract draft was wrong, not the code
jiayuww/SpatialEval/inference_vlm.py
pointer only (licence: NONE) · get_code("7dfcd86ef6835d64")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, jiayuww/SpatialEval when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

format_bunny_vqa_prompt_hf Ran
ran; our contract draft was wrong, not the code
jiayuww/SpatialEval/inference_vlm.py
pointer only (licence: NONE) · get_code("7e7acafc5ee625af")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, jiayuww/SpatialEval when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

get_chunk Ran
ran; our contract draft was wrong, not the code
BAAI-DCAI/Bunny/bunny/eval/model_vqa.py
code served (permissive licence) · get_code("42a46570620cd9fa")
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) and returned list, ndim 1, dtype family int; 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 math

def split_list(lst, n):
    """Split a list into n (roughly) equal-sized chunks"""
    chunk_size = math.ceil(len(lst) / n)
    return [lst[i:i + chunk_size] for i in range(0, len(lst), chunk_size)]

def get_chunk(lst, n, k):
    chunks = split_list(lst, n)
    return chunks[k]

# Setup code
lst = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]
n = 3
k = 1

ARGS = {
    "lst": lst,
    "n": n,
    "k": k
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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 list, ndim 1, dtype family int — 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
is_none Ran
ran, violated the contract we drafted
BAAI-DCAI/Bunny/bunny/eval/model_vqa_mmbench.py
pointer only (licence: Apache-2.0); the stored copy get_code serves, from ASTRAL-Group/ASTRA, is pointer only (licence: NONE) · get_code("bae18947b56f2be1")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from ASTRAL-Group/ASTRA, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

load_yaml Ran
ran; our contract draft was wrong, not the code
BAAI-DCAI/Bunny/bunny/eval/model_vqa_cmmmu.py
code served (permissive licence) · get_code("ce93b7e192e1a8b7")
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 tempfile
import os
import yaml

# Create a temporary YAML file with valid content
temp_dir = tempfile.mkdtemp()
yaml_file_path = os.path.join(temp_dir, "test.yaml")

# Write a simple valid YAML file
yaml_content = {
    "name": "test_config",
    "value": 42,
    "nested": {
        "key": "value",
        "list": [1, 2, 3]
    }
}

with open(yaml_file_path, 'w') as f:
    yaml.dump(yaml_content, f)

ARGS = {
    "file_path": yaml_file_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
yaml — 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
parse_multi_choice_response Ran
ran; our contract draft was wrong, not the code
BAAI-DCAI/Bunny/bunny/eval/model_vqa_mmmu.py
code served (permissive licence) · get_code("f3018307f9301da1")
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 str. 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 random
import numpy as np

# Set seed for reproducibility
random.seed(42)
np.random.seed(42)

# Define the arguments for parse_multi_choice_response
ARGS = {
    "response": "The answer is (B) because it makes the most sense.",
    "all_choices": ['A', 'B', 'C', 'D'],
    "index2ans": {
        'A': 'first option',
        'B': 'second option',
        'C': 'third option',
        'D': 'fourth option'
    }
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
remove_redundancy Ran
ran; our contract draft was wrong, not the code
jiayuww/SpatialEval/evals/evaluation.py
pointer only (licence: NONE) · get_code("85353e219884c7eb")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, jiayuww/SpatialEval when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

split_list Ran
ran; our contract draft was wrong, not the code
BAAI-DCAI/Bunny/bunny/eval/model_vqa.py
code served (permissive licence) · get_code("076c252c52cbb161")
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) and returned list. 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 math

def split_list(lst, n):
    """Split a list into n (roughly) equal-sized chunks"""
    chunk_size = math.ceil(len(lst) / n)
    return [lst[i:i + chunk_size] for i in range(0, len(lst), chunk_size)]

# Setup code
lst = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]
n = 3

ARGS = {
    "lst": lst,
    "n": n
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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 list — a list 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
tokenizer_image_token Ran
ran, honoured the contract we drafted
jiayuww/SpatialEval/inference_vlm.py
pointer only (licence: NONE) · get_code("e582401378cfcafc")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, jiayuww/SpatialEval when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

load_pretrained_model Not yet run
unverified (harvested, no recorded run)
BAAI-DCAI/Bunny/bunny/model/builder.py
code served (permissive licence) · get_code("8bc914474691d84a")
How it ran

no invocation recorded.

Also present in other papers' code

Of 13 samples lifted from this paper's linked repositories, 13 carry a function identity and 0 do not. Of the 13 distinct functions with an identity, 8 are also present in other papers whose code we hold (8 with at least one paper dated earlier, 5 the same month, 6 later) and 5 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
split_list 197 other papers
60 dated earlier, 22 the same month, 108 later, 6 same year, order not determinable, 1 undated
2112.05682 (2021-12)
Self-attention Does Not Need $O(n^2)$ Memory
2306.17107 (2023-06)
LLaVAR: Enhanced Visual Instruction Tuning for Text-Rich Image Understanding
2307.10490 (2023-07)
Abusing Images and Sounds for Indirect Instruction Injection in Multi-Modal LLMs
and 194 more not named here
get_chunk 189 other papers
58 dated earlier, 21 the same month, 104 later, 6 same year, order not determinable
2112.05682 (2021-12)
Self-attention Does Not Need $O(n^2)$ Memory
2306.17107 (2023-06)
LLaVAR: Enhanced Visual Instruction Tuning for Text-Rich Image Understanding
2307.10490 (2023-07)
Abusing Images and Sounds for Indirect Instruction Injection in Multi-Modal LLMs
and 186 more not named here
is_none 83 other papers
28 dated earlier, 10 the same month, 43 later, 2 same year, order not determinable
2112.05682 (2021-12)
Self-attention Does Not Need $O(n^2)$ Memory
2308.10253 (2023-08)
StableLLaVA: Enhanced Visual Instruction Tuning with Synthesized Image-Dialogue Data
2308.13566 (2023-08)
MLLM-DataEngine: An Iterative Refinement Approach for MLLM
and 80 more not named here
parse_multi_choice_response 9 other papers
3 dated earlier, 1 the same month, 5 later
2311.10774 (2023-11)
MMC: Advancing Multimodal Chart Understanding with Large-scale Instruction Tuning
2402.11530 (2024-02)
Efficient Multimodal Learning from Data-centric Perspective
2402.14289 (2024-02)
TinyLLaVA: A Framework of Small-scale Large Multimodal Models
and 6 more not named here
load_yaml 6 other papers
1 dated earlier, 2 the same month, 3 later
2402.11530 (2024-02)
Efficient Multimodal Learning from Data-centric Perspective
2406.05343 (2024-06)
M3GIA: A Cognition Inspired Multilingual and Multimodal General Intelligence Ability Benchmark
2406.13642 (2024-06)
SpatialBot: Precise Spatial Understanding with Vision Language Models
and 3 more not named here
tokenizer_image_token 3 other papers
1 dated earlier, 2 later · 1 only as an assembly brick
2405.15683 (2024-05)
Pre-print. Under Review. VISUAL DESCRIPTION GROUNDING REDUCES HALLU-CINATIONS AND BOOSTS REASONING IN LVLMS
2409.17143 (2024-09)
Attention Prompting on Image for Large Vision-Language Models
2412.04332 (2024-12)
Liquid: Language Models are Scalable and Unified Multi-modal Generators
construct_prompt 1 other paper
1 dated earlier
2402.11530 (2024-02)
Efficient Multimodal Learning from Data-centric Perspective
load_pretrained_model 1 other paper
1 dated earlier
2402.11530 (2024-02)
Efficient Multimodal Learning from Data-centric Perspective

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

Large language models (LLMs) and vision-language models (VLMs) have demonstrated remarkable performance across a wide range of tasks and domains. Despite this promise, spatial understanding and reasoning-a fundamental component of human cognition-remains under-explored. We propose SpatialEval, a novel benchmark that covers diverse aspects of spatial reasoning such as relationship understanding, navigation, and counting. We conduct a comprehensive evaluation of competitive language and vision-language models. Our findings reveal several counter-intuitive insights that have been overlooked in the literature: (1) Spatial reasoning poses significant challenges where competitive models can fall behind random guessing; (2) Despite additional visual input, VLMs often under-perform compared to their LLM counterparts; (3) When both textual and visual information is available, multi-modal language models become less reliant on visual information if sufficient textual clues are provided. Additionally, we demonstrate that leveraging redundancy between vision and text can significantly enhance model performance. We hope our study will inform the development of multimodal models to improve spatial intelligence and further close the gap with human intelligence. Our code is available at https://github.com/jiayuww/SpatialEval.

For agents

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

get_harvested_code_for_paper("2406.14852")
get_code_for_paper("2406.14852")
have("2406.14852")

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

curl https://syntology.ai/api/ran/2406.14852.json

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

[![Syntology run record](https://syntology.ai/api/ran/2406.14852.svg)](https://syntology.ai/paper/2406.14852)

Connect an agent — have() is free.