6 papers
SafePyramid: A Hierarchical Benchmark for In-context Policy Guardrailing
Jiacheng Zhang, Haoyu He, Sen Zhang +5
In real-world applications, guardrails are often expected to identify unsafe user-model interactions according to application-specific safety policies, rather than relying on prede…
PhyAVBench: A Challenging Audio Physics-Sensitivity Benchmark for Physically Grounded Text-to-Audio-Video Generation
Tianxin Xie, Wentao Lei, Kai Jiang +27
Text-to-audio-video (T2AV) generation is central to applications such as filmmaking and world modeling. However, current models often fail to produce physically plausible sounds. P…
NeXT-IMDL: Build Benchmark for NeXT-Generation Image Manipulation Detection & Localization
Yifei Li, Haoyuan He, Yu Zheng +5
The accessibility surge and abuse risks of user-friendly image editing models have created an urgent need for generalizable, up-to-date methods for Image Manipulation Detection and…
MemoryGraft: Persistent Compromise of LLM Agents via Poisoned Experience Retrieval
Saksham Sahai Srivastava, Haoyu He
Large Language Model (LLM) agents increasingly rely on long-term memory and Retrieval-Augmented Generation (RAG) to persist experiences and refine future performance. While this ex…
MDPO: Overcoming the Training-Inference Divide of Masked Diffusion Language Models
Haoyu He, Katrin Renz, Yong Cao +1
Diffusion language models, as a promising alternative to traditional autoregressive (AR) models, enable faster generation and richer conditioning on bidirectional context. However,…
Scholar Inbox: Personalized Paper Recommendations for Scientists
Markus Flicke, Glenn Angrabeit, Madhav Iyengar +10
Scholar Inbox is a new open-access platform designed to address the challenges researchers face in staying current with the rapidly expanding volume of scientific literature. We pr…