Hong Kong

Xiang Zheng

Advanced
@x-zheng16

Research Assistant Professor@CityUHK

x-zheng16.github.io

Awesome-Embodied-AI-Safety. Safety in Embodied AI: A Survey of Risks, Attacks, and Defenses | 500+ Papers | Perception, Cognition, Planning, Interaction, Agentic System

119

JustAsk. [ICML 2026] JustAsk: Curious Code Agents Reveal System Prompts in Frontier LLMs | Verified on Claude Code | Autoresearch for System Prompt Extraction

57

System-Prompt-Open. Open database of system prompts extracted from frontier LLMs using JustAsk

37

IMAP. [DSN 2024] Toward Evaluating Robustness of Reinforcement Learning with Adversarial Policy

8

uclaw. Turn your Claude Code into OpenClaw — bridge Telegram and Feishu to Claude Code via Claude Agent SDK

8

OpenRedRL. [FCS] OpenRedRL: A Light-Weight Benchmark for RL-Based Red Teaming

7

CALM. [AAAI 25] CALM: Curiosity-Driven Auditing for LLMs

5

CIM. [IJCAI 2024] Constrained Intrinsic Motivation for RL

4

CS5293-Tutorials. Teaching Assistant Code for CS5293 Tutorials at CityU

4

BlueSuffix. [ICLR 2025] Reinforced Blue Teaming for VLMs Against Jailbreak Attacks

1

cc-sciencepal. Claude Code plugin for SciencePal science research agents

1

cc-swarm. Peer-mesh multi-agent coordination for Claude Code over tmux

1
12
Apply