Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

HydraPrompt: An Adaptive and Asymmetric Framework of Vision-Language Models for Synthetic Image Detection

About

The rapid evolution of generative models has precipitated a proliferation of fabricated content, posing significant challenges to existing Synthetic Image Detection (SID) methods. Capitalizing on advancements in vision-language models (e.g., CLIP), recent attempts have leveraged learnable textual prompts to identify synthetic images. However, they still leverage static prompt as a fixed boundary for real and fake images, failing to adapt to the varying types of forgery that emerge during inference. To overcome this issue, we propose **HydraPrompt**, an asymmetric prompting framework that dynamically adjusts the category centers by aligning with fine-grained image cues. Specifically, we propose an Asymmetric Prompt Adapter (**APA**): (1) for authentic category, we introduce a single set of prompts to capture the consistent representative patterns, which serves as a unified anchor for real content. While (2) for fake category, we construct sample-adaptive prompts that specialize in capturing diverse cues from different samples, enabling adaptive modeling of forgery image variations. To increase pronounced discriminability within different synthetic images, we further introduce a Conditional Supervised Contrastive (**CSC**) objective, which compacts the authentic representations while capturing fine-grained forgery clues. Extensive experiments on popular SID benchmarks demonstrate the state-of-the-art performance of our framework.

Senyuan Shi, Hao Tan, Zichang Tan, Shuhan Feng, Ajian Liu, Sergio Escalera, Jun Wan• 2026

Related benchmarks

TaskDatasetResultRank
AI-generated image detectionChameleon (test)
Accuracy69.7
109
AI-generated image detectionWildRF Reddit (test)
Accuracy95.3
19
AI-generated image detectionWildRF (Facebook) (test)
Accuracy95.2
19
AI-generated image detectionWildRF Twitter (test)
Accuracy97.3
19
Synthetic Image DetectionUniversalFakeDetect Guided 49 (test)
Accuracy89.5
12
Synthetic Image DetectionUniversalFakeDetect LDM 200 steps 49 (test)
Accuracy99.5
12
Synthetic Image DetectionUniversalFakeDetect LDM 100 steps 49 (test)
Accuracy99.6
12
Synthetic Image DetectionUniversalFakeDetect Mean 49 (test)
Accuracy95.9
12
Synthetic Image DetectionUniversalFakeDetect DALL-E 49 (test)
Accuracy98.4
12
Synthetic Image DetectionUniversalFakeDetect LDM 200 w/cfg 49 (test)
Accuracy97.3
12
Showing 10 of 15 rows

Other info

Follow for update