SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2301.13741 · ICML · 2023

UPop: Unified and Progressive Pruning for Compressing Vision-Language Transformers

Ying Jin, Chun Yuan, Jiaqi Wang, Dachuan Shi, Chaofan Tao, Zhendong Yang

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 2 functions out of this paper's own repositories and ran 0 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
sdc17/crossget — 0 of 1
sdc17/upop — 0 of 1
FunctionStatusWhere it lives
TokenReduction Not yet run sdc17/crossget/LLaVA/llava/model/reducer.py
code served (permissive licence) · get_code("c5ba355bd1c09ea4")
update_alpha_parameters Not yet run sdc17/upop/compress_retrieval.py
code served (permissive licence) · get_code("f1ad8af135d8bc70")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Real-world data contains a vast amount of multimodal information, among which vision and language are the two most representative modalities. Moreover, increasingly heavier models, e.g., Transformers, have attracted the attention of researchers to model compression. However, how to compress multimodal models, especially visonlanguage Transformers, is still under-explored. This paper proposes the Unified and Progressive Pruning (UPop) as a universal vison-language Transformer compression framework, which incorporates 1) unifiedly searching multimodal subnets in a continuous optimization space from the original model, which enables automatic assignment of pruning ratios among compressible modalities and structures; 2) progressively searching and retraining the subnet, which maintains convergence between the search and retrain to attain higher compression ratios. Experiments on various tasks, datasets, and model architectures demonstrate the effectiveness and versatility of the proposed UPop framework. The code is available at https://github.com/sdc17/UPop.

For agents

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

get_harvested_code_for_paper("2301.13741")
get_code_for_paper("2301.13741")
have("2301.13741")

Connect an agent — have() is free.