Dong Li

The paper proposes AEGIS, a novel diffusion-guided method for injecting adversarial perturbations into the latent space to create generalizable and robust defenses against advanced facial deepfake manipulations.

DeepGuard introduces a novel multi-layer semantic aggregation framework to enhance secure code generation by collecting vulnerability cues from multiple upper layers of LLMs, significantly improving security while maintaining functional correctness.

The paper proposes DAMPER, a domain-aware framework that autonomously extracts and rewrites private information from text while providing rigorous differential privacy guarantees, significantly improving the privacy-utility trade-off.

The paper introduces SecGoal, a benchmark dataset and framework, demonstrating that fine-tuning smaller LLMs on this dataset significantly improves the precision of extracting formalizable security goals from natural language protocol documents.

The paper proposes SRTJ, a Self-Evolving Rule-Driven Training-Free Jailbreak framework that systematically discovers and refines attack strategies using rule composition and feedback to achieve robust and generalizable jailbreaking against modern LLMs.

This paper repurposes the statistical signals from data-poisoning backdoor attacks on contrastive learning (CL) models to create a multi-level, effective watermarking scheme for dataset intellectual property (IP) protection.

The paper introduces a framework using the 'behavioral geometry' of model populations to efficiently predict jailbreak susceptibility and transfer defenses, achieving high accuracy with significantly fewer evaluations.

C-MIG is a novel retrieval-augmented generation framework that uses multi-view information gain to improve clinical diagnosis reasoning by providing richer, more nuanced reward signals than existing methods.

The paper proposes EAPO, an entropy-driven adaptive weighting method that dynamically adjusts the influence of positive samples during policy optimization to improve both response diversity and stability in open-ended QA.

The paper introduces AgentSchool, an advanced LLM-powered multi-agent simulator that models learning as state transitions to provide a robust, ethically viable testbed for educational research and pedagogical reform.

The paper proposes a novel trace-aware decoding framework, combining Temporal-Spatial Parallel Decoding (TSPD) and Confidence Extrapolation (CE), to significantly accelerate the inference of diffusion-based LLMs by identifying and fixing converged tokens early.

The paper analyzes the entropy dynamics of Chain-of-Thought (CoT) reasoning, identifying a transition from an exploratory Uncertainty Region to a stable Confidence Region, which enables superior early exit and test-time scaling strategies.