Paper page - EviRank: Structured Relevance Evidence for Multimodal Image Re-ranking

Source: https://huggingface.co/papers/2608.20886 Authors:

,

,

,

,

,

,

,

,

,

Abstract

EviRank reformulates multimodal image re-ranking as semantic constraint satisfaction by parsing queries into structured evidence packages and verifying candidates via rubric scoring and listwise comparison without training.

Real-world image search queries are multimodal and compositional: ``find this shirt in pink’’ specifies an entity to retain, an attribute to modify, and context to ignore. Yet existing re-rankers either compress such multifaceted relevance into an opaque embedding or rely on free-form chain-of-thought that easily omits or hallucinates fine-grained constraints. Drawing on rubric- and checklist-based evaluation from NLP, we recastmultimodal image re-rankingas asemantic constraint satisfactionproblem and propose EviRank, which parses any query - text-only, image-only, or composed - into a unifiedevidence package: typed criteria across six semantic slots (e.g., entities, attributes, relations), each labelled required, forbidden, or ignorable. Re-ranking then reduces to evidence-conditioned verification, combining deterministicrubric scoringand evidence-groundedlistwise comparisonin a singletraining-freeprocedure. The explicit evidence can further serve as structured supervision for optionally distilling a lightweight student. Across five benchmarks spanning text-to-image, image-to-image, and composed image retrieval, EviRank achieves state-of-the-art performance, and thedistilled studentpreserves over 90% of the teacher’s capability at substantially lower cost.

View arXiv pageView PDFAdd to collection

Get this paper in your agent:

hf papers read 2608\.20886

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2608.20886 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2608.20886 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2608.20886 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.