Institution profile

Hohai University

Academic institutionasia · cn
Official website
Research library113linked papers
Opportunities0open roles
Selected work

Representative Papers

Diff-PC: Identity-preserving and 3D-aware controllable diffusion for zero-shot portrait customization

Dec 01, 2024Information Fusion

Existing portrait customization methods often struggle to simultaneously preserve identity fidelity and achieve precise facial control. To address this limitation, this work proposes Diff-PC, a novel framework that leverages a 3D face-guided identity encoder and a feature injection mechanism to enable high-fidelity, fine-grained controllable portrait generation under zero-shot conditions. The approach effectively integrates 3D-aware priors with identity features and is compatible with diverse backgrounds and multi-style base models. Trained with a dedicated identity-centric dataset and enhanced by an ID-Encoder, an ID-Ctrl alignment module, and an ID-Injector refinement module, Diff-PC consistently outperforms state-of-the-art methods in terms of identity preservation, facial controllability, and text-to-image alignment.

6 citationsRead paper

CodeOCR: On the Effectiveness of Vision Language Models in Code Understanding

Feb 02, 2026

This work addresses the computational burden imposed by long text sequences in large language models for code understanding. It presents the first systematic exploration of rendering source code as images and feeding them into multimodal large language models, leveraging the compressibility of visual representations to substantially reduce context length and computational cost while preserving essential semantic information. By incorporating visual cues such as syntax highlighting, the approach achieves up to an 8× token compression ratio on code completion tasks and even outperforms the original text-based input at a 4× compression rate. Furthermore, it demonstrates robust or slightly superior performance on code clone detection, validating the efficacy and potential of image-based modalities for efficient code understanding.

1 citationsRead paper

A Hybrid Autoencoder-Transformer Model for Robust Day-Ahead Electricity Price Forecasting under Extreme Conditions

Apr 25, 20252025 IEEE International Symposium on the Application of Artificial Intelligence in Electrical Engineering (AAIEE)

To address the low accuracy and poor robustness of day-ahead electricity price forecasting (DAEPF) under extreme weather and market anomalies, this paper proposes a Distillation-Attention Transformer with Autoencoder-based Self-Regression (DAT-ASR) framework. The method innovatively integrates a dynamic attention weight allocation mechanism and an unsupervised anomaly pattern separation module to jointly capture long- and short-term price dependencies while explicitly modeling anomalous disturbances. Furthermore, self-supervised pretraining enhances generalization in low-data anomaly scenarios. Experiments on California ISO and Shandong Power Market datasets demonstrate that DAT-ASR achieves 12.7%–18.3% lower average MAE than state-of-the-art models, reduces prediction errors during anomalous periods by over 25%, and improves inference speed by approximately 40%. These results substantiate significant gains in DAEPF accuracy, robustness, and practicality under extreme conditions.

1 citationsRead paper
Recent publications

Latest Papers