Composed Image Retrieval (CIR) aims to retrieve a target image given a reference image and a modification text. Although recent approaches leveraging Multimodal Large Language Models (MLLMs) have show
FactEmbed improves composed image retrieval by decoupling query understanding from embedding construction through factorized latents that simulate visual modifications before retrieval.
Composed Image Retrieval (CIR) aims to retrieve a target image given a reference image and a modification text. Although recent approaches leveraging Multimodal Large Language Models (MLLMs) have shown significant progress, we argue that the MLLM's compositional understanding 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, we introduce Factorized MLLM Embeddings (FactEmbed), which decouples compositional query comprehension from building discriminative embeddings. Rather than directly projecting the composed query into a single…
Summary and abstract excerpt via Hugging Face Papers ↗ · arXiv.
Factorized MLLM Embeddings for Composed Image Retrieval was first seen on Hugging Face · Daily papers on Nov 4, 2026. Since then it has appeared in 20 checks on 2 different days. Its best position was #3 of 30, reached on Nov 5, 07:28 UTC. It dropped off the list after Nov 5, 23:28 UTC.
Position on Hugging Face · Daily papers at every check (lower is better).
Best rank per day (UTC).
Platform metric as reported by Hugging Face.
Archived daily, weekly and monthly lists that include this entry.
Not at the moment. It was last on the list on Nov 5, 23:28 UTC, after 2 days of trending.
#3 on Hugging Face · Daily papers, on Nov 5, 07:28 UTC. Trendoday checks this list every 120 minutes.
It was first seen on Nov 4, 2026 and has been on the list on 2 days (20 checks) since.