Yi Wu

This paper systematically analyzes the resilience of LLM-enhanced search engines against black-hat SEO attacks, finding that while they block most traditional attacks, they remain vulnerable to sophisticated LLM-generated query manipulations.

The paper proposes using In-Context Learning (ICL) as a unified, proactive framework for detecting evolving illicit online promotions, achieving high accuracy with significantly fewer labeled examples and discovering novel threats.

FuzzAgent introduces a multi-agent, evolutionary system that significantly improves library fuzzing by iteratively refining the test suite based on runtime feedback, achieving superior coverage and bug detection compared to state-of-the-art methods.

The paper introduces MIRA, a bilingual benchmark that reveals that LLMs tend to dilute or omit critical medical information when responding to prompts from users with low health literacy, a pattern termed Differential Information Dilution (DID).

Real2SAM2Real introduces a framework that uses explicit 3D caches, derived from 3D lifting models, to provide robust geometric guidance to Video Diffusion Models, significantly improving spatiotemporal consistency during complex movements and occlusions.

The paper proposes a Variational Adapter (VACSR) to improve cross-modal similarity representation by treating fine-grained image-text matching as a variational inference problem, thereby mitigating the negative effects of binary annotation compression.

GRKV introduces a training-free KV-cache merging method that uses global regression to distribute information from evicted tokens, solving the over-merging problem inherent in span-based retention.

The paper proposes Deep Research as Rubric (DR-rubric), a novel evidence-driven framework that treats rubric construction itself as a research problem to generate fine-grained, scalable reward signals for open-ended reasoning tasks.