Papers › MULTIFLOW: Shifting Towards Task-Agnostic Vision-Language Pruning

MULTIFLOW: Shifting Towards Task-Agnostic Vision-Language Pruning

8 Apr 2024CVPR 2024 1arXiv:2404.05621archive 2025-07-28

Matteo Farina, Massimiliano Mancini, Elia Cunegatti, Gaowen Liu, Giovanni Iacca, Elisa Ricci

While excellent in transfer learning, Vision-Language models (VLMs) come with high computational costs due to their large number of parameters. To address this issue, removing parameters via model pruning is a viable solution. However, existing techniques for VLMs are task-specific, and thus require pruning the network from scratch for each new task of interest. In this work, we explore a new direction: Task-Agnostic Vision-Language Pruning (TA-VLP). Given a pretrained VLM, the goal is to find a unique pruned counterpart transferable to multiple unknown downstream tasks. In this challenging setting, the transferable representations already encoded in the pretrained model are a key aspect to preserve. Thus, we propose Multimodal Flow Pruning (MULTIFLOW), a first, gradient-free, pruning framework for TA-VLP where: (i) the importance of a parameter is expressed in terms of its magnitude and its information flow, by incorporating the saliency of the neurons it connects; and (ii) pruning is driven by the emergent (multimodal) distribution of the VLM parameters after pretraining. We benchmark eight state-of-the-art pruning algorithms in the context of TA-VLP, experimenting with two VLMs, three vision-language tasks, and three pruning ratios. Our experimental results show that MULTIFLOW outperforms recent sophisticated, combinatorial competitors in the vast majority of the cases, paving the way towards addressing TA-VLP. The code is publicly available at https://github.com/FarinaMatteo/multiflow.

PaperPDFConference PDFCodeCode Syntology ran

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2404.05621")

Code

Syntology Ran 8 of 27 code samples harvested from 1 repository linked to this paper; 19 have no recorded run. Of those that ran: 2 ran · honoured contract; 1 ran · violated contract; 1 ran · our draft was wrong; 4 ran with no contract checked.

By repository: official repository: 27 samples from 1 repository, 8 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

farinamatteo/multiflow officialmentioned in papermentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

27 samples harvested; 8 ran; 2 honoured the contract we drafted; 19 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

2ran · honoured contract
1ran · violated contract
1ran · our draft was wrong
4ran
19unverified

Licence: 27 of the 27 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from FarinaMatteo/multiflow. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

blip_detect_modality_fn FarinaMatteo/multiflow/pruners/multiflow.py official repository ran no licence file found · pointer only · 7e26ba779356a23e · report
is_blip_text_encoder_weight_tied FarinaMatteo/multiflow/pruners/multiflow.py official repository ran · violated contract no licence file found · pointer only · 68225426d2b546b7 · report
masks FarinaMatteo/multiflow/pruners/multiflow.py official repository ran · our draft was wrong no licence file found · pointer only · 1933ad23737e14e6 · report
recursive_getattr FarinaMatteo/multiflow/pruners/multiflow.py official repository ran · honoured contract no licence file found · pointer only · b12707e343eb192f · report
remaining_params FarinaMatteo/multiflow/pruners/multiflow.py official repository ran · honoured contract no licence file found · pointer only · 1c3162fdd29aa482 · report
vit_detect_modality_fn FarinaMatteo/multiflow/pruners/multiflow.py official repository ran fingerprinted no licence file found · pointer only · c725f131c12b446f · report
vit_forward_output FarinaMatteo/multiflow/pruners/multiflow.py official repository ran no licence file found · pointer only · 9e24be1a671d2b45 · report
xvlm_detect_modality_fn FarinaMatteo/multiflow/pruners/multiflow.py official repository ran no licence file found · pointer only · 4f9d037860e42be8 · report
GetTopSubnet FarinaMatteo/multiflow/pruners/multiflow.py official repository unverified no licence file found · pointer only · 3e0bb09ca02798d1 · report
Linear FarinaMatteo/multiflow/pruners/multiflow.py official repository unverified no licence file found · pointer only · 1de425a1a8df41d9 · report
MultiFlow FarinaMatteo/multiflow/pruners/multiflow.py official repository unverified no licence file found · pointer only · b824b71a63a9e9f9 · report
Pruner FarinaMatteo/multiflow/pruners/multiflow.py official repository unverified no licence file found · pointer only · f790d340a0df644e · report
SupermaskLinear FarinaMatteo/multiflow/pruners/multiflow.py official repository unverified no licence file found · pointer only · 2ddf606b2f63830f · report
_make_prunable FarinaMatteo/multiflow/pruners/multiflow.py official repository unverified no licence file found · pointer only · 5fc98e8ae7e46dce · report
blip_forward_output FarinaMatteo/multiflow/pruners/multiflow.py official repository unverified no licence file found · pointer only · 776ca7bcdd190f41 · report
blip_region_forward_output FarinaMatteo/multiflow/pruners/multiflow.py official repository unverified no licence file found · pointer only · ece32424a6a15c71 · report
check_blip_state_dict FarinaMatteo/multiflow/pruners/multiflow.py official repository unverified no licence file found · pointer only · 133ca3834d81b322 · report
detect_modality_fn FarinaMatteo/multiflow/pruners/multiflow.py official repository unverified no licence file found · pointer only · 773c668c55aa5c55 · report
forward_output FarinaMatteo/multiflow/pruners/multiflow.py official repository unverified no licence file found · pointer only · cda794b0932e7cd0 · report
make_prunable FarinaMatteo/multiflow/pruners/multiflow.py official repository unverified no licence file found · pointer only · e603263fe94b6fd9 · report
memorization_hook FarinaMatteo/multiflow/pruners/multiflow.py official repository unverified no licence file found · pointer only · cad07f042e2c897d · report
named_masked_parameters FarinaMatteo/multiflow/pruners/multiflow.py official repository unverified no licence file found · pointer only · 2a279264408cfa04 · report
prunable FarinaMatteo/multiflow/pruners/multiflow.py official repository unverified no licence file found · pointer only · 9b2470bea1788960 · report
region_forward_output FarinaMatteo/multiflow/pruners/multiflow.py official repository unverified no licence file found · pointer only · dcc2f99144d98cd2 · report
stats FarinaMatteo/multiflow/pruners/multiflow.py official repository unverified no licence file found · pointer only · 2d86258221d323a2 · report
xvlm_forward_output FarinaMatteo/multiflow/pruners/multiflow.py official repository unverified no licence file found · pointer only · 6b27647f011012d3 · report
xvlm_region_forward_output FarinaMatteo/multiflow/pruners/multiflow.py official repository unverified no licence file found · pointer only · 88cc4d1cf6ca542b · report

Tasks

Transfer Learning

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Pruning

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections