Selected work

Representative Papers

AI-Generated Image Detectors Overrely on Global Artifacts: Evidence from Inpainting Exchange

Jan 30, 2026

Current AI-generated image detectors exhibit insufficient reliability in identifying localized inpainting, as they overly rely on global spectral artifacts introduced by VAE reconstruction rather than on genuine synthetic content. To address this, this work proposes the Inpainting Exchange (INP-X) operation, which preserves the inpainted region’s content while restoring original pixels in non-edited areas, thereby systematically revealing— for the first time—the detectors’ dependence on global artifacts. A newly constructed 90K test set based on INP-X demonstrates a significant performance drop in state-of-the-art detectors (e.g., from 91% to 55% accuracy), approaching random guessing. Conversely, models trained on this dataset show markedly improved generalization and localization capabilities, advancing the development of content-aware detection methodologies.

0 citationsRead paper

STAMP: Spatial-Temporal Adapter with Multi-Head Pooling

Nov 13, 2025

Prior work lacks a systematic comparison of EEG-specific foundation models (EEGFMs) versus general-purpose time-series foundation models (TSFMs) on EEG analysis tasks. Method: We propose STAMP, a lightweight spatial-temporal adapter that integrates EEG’s inter-channel spatial dependencies and temporal dynamics implicitly—without modifying pretrained TSFM parameters—via univariate embedding and multi-head pooling. Contribution/Results: This is the first comprehensive evaluation of TSFMs’ applicability to EEG analysis, demonstrating their competitiveness with dedicated EEGFMs. STAMP achieves state-of-the-art performance across eight clinical EEG classification benchmarks, with significantly reduced parameter count and training cost. It enables efficient, parameter-efficient transfer of general-purpose TSFMs to EEG domains, advancing scalable and adaptable neurophysiological signal modeling.

0 citationsRead paper

A Human-Centered Approach to Identifying Promises, Risks, & Challenges of Text-to-Image Generative AI in Radiology

Jul 21, 2025

Current text-to-image generative AI in radiology lacks systematic evaluation of clinical utility and human factors integration, risking technological misalignment with real-world needs and potential safety hazards. Method: This study uniquely embeds radiologists, residents, and medical students throughout the generative AI development lifecycle, integrating a text-to-CT synthesis model, qualitative user studies, human-computer interaction analysis, and prompt engineering experiments. Contribution/Results: We identify domain-specific challenges in safety, interpretability, and accountability; delineate high-value educational applications; and propose empirically grounded design principles to mitigate misuse of synthetic medical images. Our work establishes a methodological framework and practical guidelines for responsible, clinically aligned generative AI in radiology.

0 citationsRead paper

Recent publications

Latest Papers

AI-Generated Image Detectors Overrely on Global Artifacts: Evidence from Inpainting Exchange

Jan 30, 2026

Current AI-generated image detectors exhibit insufficient reliability in identifying localized inpainting, as they overly rely on global spectral artifacts introduced by VAE reconstruction rather than on genuine synthetic content. To address this, this work proposes the Inpainting Exchange (INP-X) operation, which preserves the inpainted region’s content while restoring original pixels in non-edited areas, thereby systematically revealing— for the first time—the detectors’ dependence on global artifacts. A newly constructed 90K test set based on INP-X demonstrates a significant performance drop in state-of-the-art detectors (e.g., from 91% to 55% accuracy), approaching random guessing. Conversely, models trained on this dataset show markedly improved generalization and localization capabilities, advancing the development of content-aware detection methodologies.

0 citationsRead paper

STAMP: Spatial-Temporal Adapter with Multi-Head Pooling

Nov 13, 2025

Prior work lacks a systematic comparison of EEG-specific foundation models (EEGFMs) versus general-purpose time-series foundation models (TSFMs) on EEG analysis tasks. Method: We propose STAMP, a lightweight spatial-temporal adapter that integrates EEG’s inter-channel spatial dependencies and temporal dynamics implicitly—without modifying pretrained TSFM parameters—via univariate embedding and multi-head pooling. Contribution/Results: This is the first comprehensive evaluation of TSFMs’ applicability to EEG analysis, demonstrating their competitiveness with dedicated EEGFMs. STAMP achieves state-of-the-art performance across eight clinical EEG classification benchmarks, with significantly reduced parameter count and training cost. It enables efficient, parameter-efficient transfer of general-purpose TSFMs to EEG domains, advancing scalable and adaptable neurophysiological signal modeling.

0 citationsRead paper

A Human-Centered Approach to Identifying Promises, Risks, & Challenges of Text-to-Image Generative AI in Radiology

Jul 21, 2025

Current text-to-image generative AI in radiology lacks systematic evaluation of clinical utility and human factors integration, risking technological misalignment with real-world needs and potential safety hazards. Method: This study uniquely embeds radiologists, residents, and medical students throughout the generative AI development lifecycle, integrating a text-to-CT synthesis model, qualitative user studies, human-computer interaction analysis, and prompt engineering experiments. Contribution/Results: We identify domain-specific challenges in safety, interpretability, and accountability; delineate high-value educational applications; and propose empirically grounded design principles to mitigate misuse of synthetic medical images. Our work establishes a methodological framework and practical guidelines for responsible, clinically aligned generative AI in radiology.

0 citationsRead paper