Xu Li

The paper proposes UniRule, a novel agentic RAG framework that unifies the detection rule generation process by mapping context and language to rules, significantly outperforming pure LLM generation.

The paper introduces MGTEVAL, a comprehensive and extensible platform designed to systematically evaluate the performance, robustness, and efficiency of machine-generated text detectors.

The paper introduces FragBench, a novel benchmark designed to detect malicious LLM attacks that are split across multiple, seemingly benign sessions, showing that cross-session graph modeling is necessary for effective defense.

The paper introduces Thinking as Compression (TaC), a novel paradigm showing that the inherent reasoning process of a large language model can naturally compress long context inputs, outperforming dedicated compression methods.

The paper addresses the 'detection-to-abstention gap' in reasoning models, where detecting insufficient information does not lead to abstention, by proposing a novel control framework that forces models to commit to an answerability judgment before solving.

The paper introduces MTAVG-Bench 2.0, a new benchmark designed to diagnose high-level failure modes of cinematic expressiveness in multi-talker audio-video generation, showing that even advanced models struggle with complex scene-level failures.

EvoGens is an evolution-inspired framework that treats scientific idea generation as an evolutionary search, significantly boosting the novelty and diversity of generated research ideas compared to existing LLM-based methods.

TrafficRAG is a multimodal retrieval-augmented framework that automates traffic accident liability determination by integrating visual evidence, structured legal knowledge, and advanced LLM reasoning.