HyperAlign: Hyperbolic Entailment Cones for Adaptive Text-to-Image Alignment Assessment

TL;DR

HyperAlign introduces a hyperbolic geometric framework for more accurate and adaptive assessment of text-to-image alignment, surpassing Euclidean-based methods in evaluation tasks.

Contribution

It proposes a novel hyperbolic entailment geometry approach with dynamic supervision and adaptive modulation for improved alignment assessment.

Findings

Achieves state-of-the-art performance on evaluation benchmarks.

Demonstrates strong cross-database generalization.

Validates effectiveness of hyperbolic modeling for semantic alignment.

Abstract

With the rapid development of text-to-image generation technology, accurately assessing the alignment between generated images and text prompts has become a critical challenge. Existing methods rely on Euclidean space metrics, neglecting the structured nature of semantic alignment, while lacking adaptive capabilities for different samples. To address these limitations, we propose HyperAlign, an adaptive text-to-image alignment assessment framework based on hyperbolic entailment geometry. First, we extract Euclidean features using CLIP and map them to hyperbolic space. Second, we design a dynamic-supervision entailment modeling mechanism that transforms discrete entailment logic into continuous geometric structure supervision. Finally, we propose an adaptive modulation regressor that utilizes hyperbolic geometric features to generate sample-level modulation parameters, adaptively calibrating Euclidean cosine similarity to predict the final score. HyperAlign achieves highly competitive performance on both single database evaluation and cross-database generalization tasks, fully validating the effectiveness of hyperbolic geometric modeling for image-text alignment assessment.