pith. sign in

arxiv: 2507.08064 · v3 · pith:LJBX42KQnew · submitted 2025-07-10 · 💻 cs.MM · cs.CV

PUMA: Layer-Pruned Language Model for Efficient Unified Multimodal Retrieval with Modality-Adaptive Learning

classification 💻 cs.MM cs.CV
keywords learningmultimodallanguagelayer-prunedmodality-adaptiveretrievalunifiedefficiency
0
0 comments X
read the original abstract

As multimedia content expands, the demand for unified multimodal retrieval (UMR) in real-world applications increases. Recent work leverages multimodal large language models (MLLMs) to tackle this task. However, their large parameter size results in high training costs and low inference efficiency. To address this, we propose PUMA: a Layer-Pruned Language Model for Efficient Unified Multimodal Retrieval with Modality-Adaptive Learning. Our approach improves UMR from both structural and learning perspectives. (1) Structurally, we propose Layer-Pruned Self-Distillation, which prunes MLLMs by keeping only shallow layers while distilling features from dropped deep layers as teacher signals. This reduces parameters and preserves representation capability. (2) On the learning side, we introduce Modality-Adaptive Contrastive Learning Loss (MAC-Loss), which separates in-batch negatives into harder intra-modality and easier inter-modality groups based on the target modality, assigning different temperature strategies to enhance learning efficiency. Experiments show our method significantly reduces resource usage while maintaining strong performance.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 1 Pith paper

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

  1. FreeRet: MLLMs as Training-Free Retrievers

    cs.CV 2025-09 unverdicted novelty 6.0

    FreeRet enables pretrained MLLMs to act as training-free retrievers via semantically grounded embeddings and reasoning-based reranking, outperforming models trained on millions of pairs on MMEB benchmarks.