Ke Xu

Search agents connect LLMs to the Internet, enabling them to access broader and more up-to-date information. However, this also introduces a new threat surface: unreliable search results can mislead agents into producing unsafe outputs.…

Multi-hop audio-visual reasoning remains challenging for Omni-LLMs, as relevant evidence is often sparse, temporally dispersed, and distributed across both audio and visual streams. Existing benchmarks provide limited investigation of this…

The pretraining-finetuning paradigm is a crucial strategy in metallic surface defect detection for mitigating the challenges posed by data scarcity. However, its implementation presents a critical dilemma. Pretraining on natural image…

Topological consistency plays a crucial role in the task of boundary segmentation for reticular images, such as cell membrane segmentation in neuron electron microscopic images, grain boundary segmentation in material microscopic images and…

LLM-based agents are increasingly deployed for complex tasks requiring planning, tool use, and interaction with external services. Their reliance on untrusted external content exposes them to indirect prompt injection (IPI), in which…

Recent advances in flow matching models have significantly improved text-to-image generation quality, but also introduce growing safety risks due to the generation of harmful or undesirable content. Existing concept erasure methods are…

We investigate the rest-frame optical size-stellar mass relation of galaxies at $0.5<z<6.0$ using deep JWST/NIRCam and MIRI imaging from the PRIMER survey. We find that star-forming galaxies (SFGs) exhibit a broken power-law relation at all…

Text-to-image (T2I) models can generate not-safe-for-work (NSFW) content, motivating multi-stage safety pipelines with both text and image filters. Newer LLM-based filters detect latent intent beyond keywords, making token-level…

The deployment of quantized neural networks on edge devices, combined with privacy regulations like GDPR, creates an urgent need for machine unlearning in quantized models. However, existing methods face critical challenges: they induce…

Linguistic ambiguity is critical to the robustness of Large Language Models (LLMs), yet existing research focuses mostly on English, with limited attention devoted to Chinese. Existing Chinese ambiguity datasets (e.g., CHAmbi) suffer from…

Galaxy clusters in the local Universe are dominated by massive quiescent galaxies with old ages, formed at high redshifts. It is debated whether their quenching is driven by internal processes or environmental effects, which has been…

Short-term plasticity (STP) is fundamental to temporal information processing in biological neural systems but remains difficult to realize efficiently in neuromorphic hardware. Memristive electrochemical random-access memory (ECRAM)…

We report that the quantum-confined Stark effect spectrum exhibits a nearly rigid redshift while preserving its characteristic peak spacing patterns when increasing the electric field strength F. Using InGaN as a model system, we uncover…

We present WaferSAGE, a framework for wafer defect visual question answering using small vision-language models. To address data scarcity in semiconductor manufacturing, we propose a three-stage synthesis pipeline incorporating structured…

We propose glaucoma lesion evaluation and analysis with multimodal imaging (GLEAM), the first publicly available tri-modal glaucoma dataset comprising scanning laser ophthalmoscopy fundus images, circumpapillary OCT images, and visual field…

While text-to-image models have made strong progress in visual fidelity, faithfully realizing complex visual intents remains challenging because many requirements must be tracked across grounding, generation, and verification. We refer to…

Alignment of large language models (LLMs) via SFT and RLHF/DPO typically ignores the global geometry of the representation space, relying instead on local token likelihoods or scalar scores. We view generation as tracing a semantic…

Verification presents a major bottleneck in Integrated Circuit (IC) development, consuming nearly 70% of total effort. While the Universal Verification Methodology (UVM) improves reuse through structured verification environments,…

Industrial Control Protocols (ICPs) are critical to the reliability and stability of industrial infrastructure, yet their security is fundamentally compromised by a specification-blindness bottleneck. Modern fuzzers, constrained by…

Proprietary large language models (LLMs) embody substantial economic value and are generally exposed only as black-box APIs, yet adversaries can still exploit their outputs to extract knowledge via distillation. Existing defenses focus…