FactEmbed proposed to improve composed image retrieval
FactEmbed was announced to optimize MLLM compositional understanding in CIR, which retrieves target images using a reference image and modification text.
Composed Image Retrieval (CIR) aims to retrieve a target image given a reference image and a modification text. While recent approaches leveraging Multimodal Large Language Models (MLLMs) have shown significant progress, the researchers argue that the compositional understanding of MLLMs remains underutilized.
This stems from directly mapping composed queries into a retrieval embedding via contrastive supervision alone, without explicitly resolving the intended visual modifications. In this work, the authors introduce Factorized MLLM Embeddings (FactEmbed), which decouples compositional query comprehension from building discriminative embeddings. The paper received 37 upvotes on Hugging Face Daily Papers.
This item is a summary. The full, checkable version is at the original source:
Read the original at Hugging Face Papers ↗