SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2412.03324 · 2024

A Stitch in Time Saves Nine: Small VLM is a Precise Guidance for Accelerating Large VLMs

arXiv · PDF · Open in the Atlas

Code that ran

4 of 5 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; 2 honoured the contract we drafted, 2 violated it and 0 have no contract verdict; 0 failures of our instrument; 1 no recorded run); 1 of them was also run on the host, not in the jail, by the peer-agreement lane that set its level, 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/2412.03324.json · badge: /api/ran/2412.03324.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
NUS-HPC-AI-Lab/SGL canonical (classifier label)
classifier basis: classifier_2025
0 ran · 1 no recorded run
copy not recorded community
not in the paper's declared repositories
4 ran · 0 no recorded run
FunctionStatusWhere it lives
version_cmp Ran
ran, violated the contract we drafted
this paper's copy was not recorded; identical code first harvested from sotayang/svbench
pointer only; the stored copy get_code serves, from sotayang/svbench, is pointer only (licence: NONE) · get_code("5f3bdd5f3304a0e0")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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.

evaluate_exact_match_accuracy Ran
ran, honoured the contract we drafted
this paper's copy was not recorded; identical code first harvested from brandon3964/multimodal-task-vector
pointer only; the stored copy get_code serves, from brandon3964/multimodal-task-vector, is pointer only (licence: NOASSERTION) · get_code("e159a001bd4cbfaa")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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.

evaluate_relaxed_accuracy Ran
ran, honoured the contract we drafted
this paper's copy was not recorded; identical code first harvested from brandon3964/multimodal-task-vector
pointer only; the stored copy get_code serves, from brandon3964/multimodal-task-vector, is pointer only (licence: NOASSERTION) · get_code("4ffcad52354b4806")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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.

relaxed_correctness Ran
ran, violated the contract we drafted
this paper's copy was not recorded; identical code first harvested from brandon3964/multimodal-task-vector
pointer only; the stored copy get_code serves, from brandon3964/multimodal-task-vector, is pointer only (licence: NOASSERTION) · get_code("b3c8680ebd8739af")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. 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_attention_rank Not yet run
unverified (harvested, no recorded run)
NUS-HPC-AI-Lab/SGL/internvl/model/internvl_chat/modeling_internvl_chat.py
pointer only (licence: NONE) · get_code("88a4b26411ded833")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, NUS-HPC-AI-Lab/SGL 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. No invocation recorded.

Also present in other papers' code

Of 5 samples lifted from this paper's linked repositories, 5 carry a function identity and 0 do not. Of the 5 distinct functions with an identity, 4 are also present in other papers whose code we hold (4 with at least one paper dated earlier, 4 the same month, 4 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
evaluate_relaxed_accuracy 6 other papers
4 dated earlier, 1 the same month, 1 later
2406.08487 (2024-06)
Beyond LLaVA-HD: Diving into High-Resolution Large Multimodal Models
2406.15334 (2024-06)
Multimodal Task Vectors Enable Many-Shot Multimodal In-Context Learning
2407.01976 (2024-07)
A Bounding Box is Worth One Token: Interleaving Layout and Text in a Large Language Model for Document Understanding
and 3 more not named here
relaxed_correctness 6 other papers
4 dated earlier, 1 the same month, 1 later
2406.08487 (2024-06)
Beyond LLaVA-HD: Diving into High-Resolution Large Multimodal Models
2406.15334 (2024-06)
Multimodal Task Vectors Enable Many-Shot Multimodal In-Context Learning
2407.01976 (2024-07)
A Bounding Box is Worth One Token: Interleaving Layout and Text in a Large Language Model for Document Understanding
and 3 more not named here
version_cmp 6 other papers
1 dated earlier, 1 the same month, 4 later
2410.16261 (2024-10)
Mini-InternVL: A Flexible-Transfer Pocket Multimodal Model with 5% Parameters and 90% Performance
2412.09616 (2024-12)
V2PE: Improving Multimodal Long-Context Capability of Vision-Language Models with Variable Visual Position Encoding
2501.06986 (2025-01)
Rethinking the Mixture of Vision Encoders Paradigm for Enhanced Visual Understanding in Multimodal LLMs
and 3 more not named here
evaluate_exact_match_accuracy 4 other papers
2 dated earlier, 1 the same month, 1 later
2406.15334 (2024-06)
Multimodal Task Vectors Enable Many-Shot Multimodal In-Context Learning
2410.16261 (2024-10)
Mini-InternVL: A Flexible-Transfer Pocket Multimodal Model with 5% Parameters and 90% Performance
2412.05185 (2024-12)
LinVT: Empower Your Image-level Large Language Model to Understand Videos
and 1 more not named here

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

Vision-language models (VLMs) have shown remarkable success across various multi-modal tasks, yet large VLMs encounter significant efficiency challenges due to processing numerous visual tokens. A promising approach to accelerating large VLM inference is using partial information, such as attention maps from specific layers, to assess token importance and prune less essential tokens. However, our study reveals three key insights: (i) Partial attention information is insufficient for accurately identifying critical visual tokens, resulting in suboptimal performance, especially at low token retention ratios; (ii) Global attention information, such as the attention map aggregated across all layers, more effectively preserves essential tokens and maintains comparable performance under aggressive pruning. However, the attention maps from all layers requires a full inference pass, which increases computational load and is therefore impractical in existing methods; and (iii) The global attention map aggregated from a small VLM closely resembles that of a large VLM, suggesting an efficient alternative. Based on these findings, we introduce a \textbf{training-free} method, \underline{\textbf{S}}mall VLM \underline{\textbf{G}}uidance for accelerating \underline{\textbf{L}}arge VLMs (\textbf{SGL}). Specifically, we employ the attention map aggregated from a small VLM to guide visual token pruning in a large VLM. Additionally, an early exiting mechanism is developed to fully use the small VLM's predictions, dynamically invoking the larger VLM only when necessary, yielding a superior trade-off between accuracy and computation. Extensive evaluations across 11 benchmarks demonstrate the effectiveness and generalizability of SGL, achieving up to 91\% pruning ratio for visual tokens while retaining competitive performance.

For agents

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

get_harvested_code_for_paper("2412.03324")
get_code_for_paper("2412.03324")
have("2412.03324")

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

curl https://syntology.ai/api/ran/2412.03324.json

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

[![Syntology run record](https://syntology.ai/api/ran/2412.03324.svg)](https://syntology.ai/paper/2412.03324)

Connect an agent — have() is free.