Genecis: A Benchmark For General Conditional Image Similarity

Abstract

We argue that there are many notions of 'similarity' and that models, like humans, should be able to adapt to these dynamically. This contrasts with most representation learning methods, supervised or self-supervised, which learn a fixed embedding function and hence implicitly assume a single notion of similarity. For instance, models trained on ImageNet are biased towards object categories, while a user might prefer the model to focus on colors, textures or specific elements in the scene. In this paper, we propose the GeneCIS ('genesis') benchmark, which measures models' ability to adapt to a range of similarity conditions. Extending prior work, our benchmark is designed for zero-shot evaluation only, and hence considers an open-set of similarity conditions. We find that baselines from powerful CLIP models struggle on GeneCIS and that performance on the benchmark is only weakly correlated with ImageNet accuracy, suggesting that simply scaling existing methods is not fruitful. We furth

Stats

• citations21
• S2 citations—
• github stars0
• HF likes0
• heat score10.07
• arxiv keyvaze2023genecis

Related papers

• Conditional Similarity Networks (2016)15.06
• CSIM: A Copula-based Similarity Index Sensitive To Local Changes For Image Quality Assessment (2024)2.16
• Irgen: Generative Modeling For Image Retrieval (2023)7.16
• Elvis: Efficient Visual Similarity From Local Descriptors That Generalizes Across Domains (2026)2.35
• Advancing Similarity Search With Genai: A Retrieval Augmented Generation Approach (2024)0.00
• Id-sim: An Identity-focused Similarity Metric (2026)0.00
• Represent, Compare, And Learn: A Similarity-aware Framework For Class-agnostic Counting (2022)14.58
• Corrembed: Evaluating Pre-trained Model Image Similarity Efficacy With A Novel Metric (2023)5.24