Kai Zhang

SelfGrader proposes a lightweight, robust guardrail for detecting LLM jailbreaks by formulating the detection problem as a numerical grading task using anchored token-level logits, achieving strong performance across various benchmarks.

XMark introduces a novel multi-bit watermarking technique that reliably embeds binary messages into LLM-generated text while maintaining high text quality and robust performance even with limited token context.

ClawGuard introduces a passive, out-of-band security monitor that detects LLM agent workflow hijacking by analyzing unique electromagnetic (EM) emanations generated during agent skill execution.

The paper introduces MT-JailBench, a modular framework for evaluating multi-turn jailbreaks, demonstrating that controlling experimental components like prompt generation and resource budgets is crucial for fair comparison and understanding attack success.

The paper introduces Causal Editing (CODE), a new paradigm that improves knowledge updates in LLMs by grounding fact injection in causal narratives, drastically reducing self-refutation rates.

The paper introduces CardioLens, a rigorous evaluation testbed for multi-sequence Cardiac MRI, which reveals that current Multimodal Large Language Models (MLLMs) exhibit a significant 'clinical reality gap' and perform poorly when simulating real-world cardiac interpretation workflows.

The paper introduces Humanoid-GPT, a large-scale generative Transformer model that achieves robust zero-shot motion tracking and control by training on a massive, unified corpus of motion data.

This survey provides a systematic framework and taxonomy for evidence tracing and execution provenance in LLM agents, addressing the difficulty of verifying and auditing complex agent behaviors.

This paper explores tools for control engineers to design socio-technical systems in a more principled and ethical manner, using feedback optimization, control of Markov decision processes, and model predictive control.

This paper introduces Chronos, a physics-informed framework for non-Markovian long-horizon manipulation, which elevates observation history to the latent state of the policy dynamics and achieves higher success rates and fewer parameters than Markovian VLA baselines in both simulated and real-world experiments.

A cloud-scale gateway system for MCP services is presented, which breaks the direct-connect model and offloads legacy service integration, consolidates incompatible MCP variants, and reduces tool selection time and token usage.