AdaptiveEmbed: Sample-Adaptive Multi-Vector Representation for Multimodal Retrieval
- Published
- Source
- arXiv
- Paper number
- 1021
- Field
- Computer Vision
- arXiv ID
- 2608.25412
Key points
- It motivated the problem with the observation that the optimal number of vectors varies by sample, ranging from 1 to 8.
- Multi-Group Contrastive Learning (MGCL) produces candidate vectors, and Utility Policy Optimization (UPO) determines the number of vectors for each sample.
- On COCO, it achieved the best performance in 6 of 8 directions with an average of 2.1 vectors, and also led in zero-shot retrieval on OpenImages with a gallery of 546,000 items.
- It also outperformed fixed-budget approaches in video and audio retrieval with an average of 1.9 vectors, demonstrating that the approach is modality-independent.
- An oracle allocation could reach an average of 61.0 mAP at a similar average budget, suggesting substantial potential in adaptive allocation itself.
Paper links
External research summaries. These are not HDATF publications or measured product results.