Leaner And Faster: Two-stage Model Compression For Lightweight Text-image Retrieval

Abstract

Current text-image approaches (e.g., CLIP) typically adopt dual-encoder architecture using pre-trained vision-language representation. However, these models still pose non-trivial memory requirements and substantial incremental indexing time, which makes them less practical on mobile devices. In this paper, we present an effective two-stage framework to compress large pre-trained dual-encoder for lightweight text-image retrieval. The resulting model is smaller (39% of the original), faster (1.6x/2.9x for processing image/text respectively), yet performs on par with or better than the original full model on Flickr30K and MSCOCO benchmarks. We also open-source an accompanying realistic mobile image search application.

Stats

• citations6
• S2 citations—
• github stars0
• HF likes0
• heat score6.34
• arxiv keyren2022leaner

Related papers

• Mobileclip: Fast Image-text Models Through Multi-modal Reinforced Training (2023)18.12
• Compressible And Searchable: Ai-native Multi-modal Retrieval System With Learned Image Compression (2024)0.00
• Efficient Discriminative Joint Encoders For Large Scale Vision-language Reranking (2025)0.00
• Conaclip: Exploring Distillation Of Fully-connected Knowledge Interaction Graph For Lightweight Text-image Retrieval (2023)4.52
• Lexlip: Lexicon-bottlenecked Language-image Pre-training For Large-scale Image-text Retrieval (2023)10.85
• Coarse-to-fine: Learning Compact Discriminative Representation For Single-stage Image Retrieval (2023)9.35
• Optimizing CLIP Models For Image Retrieval With Maintained Joint-embedding Alignment (2024)6.34
• Mobileviclip: An Efficient Video-text Model For Mobile Devices (2025)2.76