Pith. sign in

REVIEW 2 cited by

COMCAT: Towards Efficient Compression and Customization of Attention-Based Vision Models

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2305.17235 v2 pith:GC5J3DLZ submitted 2023-05-26 cs.CV cs.AIcs.LG

classification cs.CVcs.AIcs.LG
keywords compressionvisionmodelsefficientattention-basedmodelappliedbeen
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
abstract

Attention-based vision models, such as Vision Transformer (ViT) and its variants, have shown promising performance in various computer vision tasks. However, these emerging architectures suffer from large model sizes and high computational costs, calling for efficient model compression solutions. To date, pruning ViTs has been well studied, while other compression strategies that have been widely applied in CNN compression, e.g., model factorization, is little explored in the context of ViT compression. This paper explores an efficient method for compressing vision transformers to enrich the toolset for obtaining compact attention-based vision models. Based on the new insight on the multi-head attention layer, we develop a highly efficient ViT compression solution, which outperforms the state-of-the-art pruning methods. For compressing DeiT-small and DeiT-base models on ImageNet, our proposed approach can achieve 0.45% and 0.76% higher top-1 accuracy even with fewer parameters. Our finding can also be applied to improve the customization efficiency of text-to-image diffusion models, with much faster training (up to $2.6\times$ speedup) and lower extra storage cost (up to $1927.5\times$ reduction) than the existing works.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. DIVE: Taming DINO for Subject-Driven Video Editing

    cs.CV 2024-12 conditional novelty 6.0 of 10

    DIVE uses DINOv2 feature maps as automatic video correspondences to carry source motion, while LoRA adapters carry the target identity.

  2. Preliminary Explorations with GPT-4o(mni) Native Image Generation

    cs.CV 2025-05 conditional novelty 5.0 of 10

    A qualitative exploration showing GPT-4o image generation excels at stylization, editing, and personalization but struggles with spatial reasoning, knowledge-based accuracy, and temporal prediction.

Pith tools