2 citations · 6 across the 23 of their papers we have counts for
27 papers
Who Improves the Robot? Toward Physical Recursive Self-Improvement
Hai Zhou, Fan Yang, Zeyi Chen +27
Robots are increasingly autonomous in task execution, yet their improvement remains largely human-driven. This study investigates Physical Recursive Self-Improvement (Physical RSI)…
Groundability, Not Scale Alone: When Weak Reviewers Can Audit Strong Coding Agents
Junyu Guo, Shangding Gu, Ming Jin +1
Coding agents can return plausible patches that omit required behavior. These failures are hard to review because long traces and confident summaries often hide what was missed. We…
When Context Changes: Understanding Update Failures in LLMs
Junyu Guo, Yuchen Fang, Shangding Gu +3
As preferences, goals, and facts change, LLM agents must use the current state while earlier versions remain in context. Yet they can answer with an old value of the same variable,…
TRACE: Governing Memory Validity in Evolving Multi-Agent Systems
Wenjun Xiong, Shengtao Zhang, Shangding Gu +5
Persistent memory lets language-model agents carry information across long-running collaborations, but leaves a lifecycle question open: what may a returning agent still act on onc…
WolfSociety: Understanding Collective Risk from Harmful-Agent Scaling in Financial Agent Societies
Lejun Zhang, Sarah Lu-Liang, Xin Jiang +3
Safety evaluations typically focus on individual agents, but interacting agents can spread harmful information and influence the environment in which later decisions are made. We s…
MAPLE-Guard: Memory-Aware Link Enforcement Against Memory-Link Poisoning in Multi-Agent Systems
Wenjun Xiong, Yijin Zhou, Jiaqian Wang +6
LLM-based multi-agent systems (MAS) increasingly rely on persistent private and shared memories for long-horizon coordination. This memory layer improves continuity, but it also gi…
Understanding and Evaluating Claw-like Agent Security Through a Computer-Systems Lens
Peizhi Niu, Wenjie Qu, Shangding Gu +14
Claw-like AI agents (e.g., OpenClaw) are always-on processes with persistent access to credentials, files, tools, and external services. They take on system-level responsibilities…
Agents' Last Exam
Yiyou Sun, Xinyang Han, Weichen Zhang +378
Recent AI systems have achieved strong results on a wide range of benchmarks, yet these gains have not translated into economically meaningful deployment across many professional d…
From Model Scaling to System Scaling: Scaling the Harness in Agentic AI
Shangding Gu
This paper studies the next major bottleneck in agentic AI as system scaling, not only model scaling: the design of auditable, persistent, modular, and verifiable architectures aro…
Remembering More, Risking More: Longitudinal Safety Risks in Memory-Equipped LLM Agents
Ahmad Al-Tawaha, Shangding Gu, Peizhi Niu +2
Safety evaluations of memory-equipped LLM agents typically measure within-task safety: whether an agent completes a single scenario safely, often under adversarial conditions such…
LLMs Should Express Uncertainty Explicitly
Junyu Guo, Shangding Gu, Ming Jin +2
Large language models (LLMs) often produce confident yet incorrect answers, which can lead to risky failures in real-world applications. We study whether post-training can make a m…
Long Context, Less Focus: A Scaling Gap in LLMs Revealed through Privacy and Personalization
Shangding Gu
Large language models (LLMs) are increasingly deployed in privacy-critical and personalization-oriented scenarios, yet the role of context length in shaping privacy leakage and per…
AgenticPay: A Multi-Agent LLM Negotiation System for Buyer-Seller Transactions
Xianyang Liu, Shangding Gu, Dawn Song
Large language model (LLM)-based agents are increasingly expected to negotiate, coordinate, and transact autonomously, yet existing benchmarks lack principled settings for evaluati…
Understanding Agent Scaling in LLM-Based Multi-Agent Systems via Diversity
Yingxuan Yang, Chengrui Qu, Muning Wen +5
LLM-based multi-agent systems (MAS) have emerged as a promising approach to tackle complex tasks that are difficult for individual LLMs. A natural strategy is to scale performance…
AccidentBench: Benchmarking Multimodal Understanding and Reasoning in Vehicle Accidents and Beyond
Shangding Gu, Xiaohan Wang, Donghao Ying +9
Rapid advances in multimodal models demand benchmarks that rigorously evaluate understanding and reasoning in safety-critical, dynamic real-world settings. We present AccidentBench…
StyleBench: Evaluating thinking styles in Large Language Models
Junyu Guo, Shangding Gu, Ming Jin +2
Structured reasoning can improve the inference performance of large language models (LLMs), but it also introduces computational cost and control constraints. When additional reaso…
Agentic Web: Weaving the Next Web with AI Agents
Yingxuan Yang, Mulei Ma, Yuxuan Huang +15
The emergence of AI agents powered by large language models (LLMs) marks a pivotal shift toward the Agentic Web, a new phase of the internet defined by autonomous, goal-driven inte…
Few-Shot Test-Time Optimization Without Retraining for Semiconductor Recipe Generation and Beyond
Shangding Gu, Donghao Ying, Ming Jin +4
We introduce Model Feedback Learning (MFL), a novel test-time optimization framework for optimizing inputs to pre-trained AI models or deployed hardware systems without requiring a…
Safe Continual Domain Adaptation after Sim2Real Transfer of Reinforcement Learning Policies in Robotics
Josip Josifovski, Shangding Gu, Mohammadhossein Malmir +5
Domain randomization has emerged as a fundamental technique in reinforcement learning (RL) to facilitate the transfer of policies from simulation to real-world robotic applications…
Robust Gymnasium: A Unified Modular Benchmark for Robust Reinforcement Learning
Shangding Gu, Laixi Shi, Muning Wen +5
Driven by inherent uncertainty and the sim-to-real gap, robust reinforcement learning (RL) seeks to improve resilience against the complexity and variability in agent-environment s…
Don't Trade Off Safety: Diffusion Regularization for Constrained Offline RL
Junyu Guo, Zhi Zheng, Donghao Ying +4
Constrained reinforcement learning (RL) seeks high-performance policies under safety constraints. We focus on an offline setting where the agent has only a fixed dataset -- common…
Enhancing Efficiency of Safe Reinforcement Learning via Sample Manipulation
Shangding Gu, Laixi Shi, Yuhao Ding +4
Safe reinforcement learning (RL) is crucial for deploying RL agents in real-world applications, as it aims to maximize long-term rewards while satisfying safety constraints. Howeve…
Safe Multi-Agent Reinforcement Learning with Bilevel Optimization in Autonomous Driving
Zhi Zheng, Shangding Gu
Ensuring safety in MARL, particularly when deploying it in real-world applications such as autonomous driving, emerges as a critical challenge. To address this challenge, tradition…
Safe and Balanced: A Framework for Constrained Multi-Objective Reinforcement Learning
Shangding Gu, Bilgehan Sel, Yuhao Ding +4
In numerous reinforcement learning (RL) problems involving safety-critical systems, a key challenge lies in balancing multiple objectives while simultaneously meeting all stringent…