Jie Wang

The paper proposes TRIP-RAG, a dynamic anonymization framework that selectively anonymizes sensitive entities in knowledge bases used for RAG, significantly improving utility while maintaining strong privacy protection.

The paper introduces AutoEG, a fully automated multi-agent framework that significantly improves the exploitation of known third-party vulnerabilities in black-box web applications by achieving an 82.41% average success rate.

OrchJail introduces an orchestration-guided fuzzing framework to systematically jailbreak tool-calling text-to-image agents by exploiting unsafe multi-step tool-orchestration patterns.

The paper proposes EVA, a novel framework that uses direct model editing to surgically correct specific neurons responsible for jailbreaking vulnerabilities in LLMs and VLMs, achieving robust safety alignment without performance degradation.

The paper proposes FRA-Attack, a frequency-domain regularization method, to significantly improve the transferability of adversarial attacks against closed-source Multimodal Large Language Models (MLLMs).

This paper introduces the concept of 'Sleeper Attack,' demonstrating that adversarial content can persist across multiple interactions with an LLM agent, posing a more subtle and difficult-to-detect safety threat than single-interaction attacks.

MIRA proposes a novel source-aware filtering framework that discovers and anchors evaluation rubrics during data selection, significantly improving code-oriented mid-training data quality while reducing token usage.

The paper introduces AgentDoG 1.5, a lightweight and scalable alignment framework that significantly improves AI agent safety and security for complex, open-world agentic scenarios.

The paper introduces Opt-Verifier, a novel LLM-based framework that significantly improves the accuracy of automated optimization model generation by implementing dual-side verification from both structural and solution perspectives.

The paper introduces a novel, training-free method to automatically generate fine-grained evaluation rubrics for LLM-as-a-Judge, and further proposes an iterative fine-tuning strategy that significantly improves rubric quality.

The paper introduces AgentDoG 1.5, a lightweight and scalable alignment framework that significantly improves AI agent safety and security for complex open-world agent deployments.

The paper proposes DARTS, a distribution-aware active rollout trajectory shaping method that fundamentally accelerates LLM reinforcement learning by actively shaping the long-tail response distribution towards conciseness and certainty.

The paper proposes S2L-PO, a framework that uses smaller, naturally diverse models as structured explorers to enhance the policy-level diversity and performance of larger language models during training.

FALAT is a diagnostic framework that treats failure attribution in complex LLM agent trajectories as a dependency-guided search problem, successfully identifying both the responsible agent and the decisive failure step.

The paper introduces Temperature-Scaled On-Policy Self-Distillation (TS-OPSD), a novel method that internalizes temperature-based policy reheating into model parameters to combat entropy collapse in reinforcement learning.

This paper introduces personalized empathy, a capability for LLMs to adapt empathetic strategies based on individual user history, and proposes PereGRM, a reward modeling framework that significantly enhances this personalized empathy.

ParetoPilot introduces a novel zero-surrogate diffusion framework for offline multi-objective optimization, achieving state-of-the-art performance by directly guiding the generation process without relying on external surrogate models.

The paper introduces the concept of Search-Time Contamination (STC), demonstrating that deep research agents can leak information from public benchmarks via web search, leading to an overestimation of their true reasoning ability.

The paper proposes OneReason, a framework that enhances the reasoning capability of generative recommendation models by focusing on improving item perception and structuring user behavior into coherent latent interests.

This paper presents EurekAgent, an environment-engineered agent system for metric-driven autonomous scientific discovery.