9 papers
Credit Without Ground Truth: Claim-Matched Evaluation of Step-Level Credit with Finite Executed Replay
Haiyue Zhang
Step-level credit evaluation can conflate numerical association with ordering and branch recovery with position-selection value. We test these distinctions in ALFWorld with two ind…
WeClawArena: An Auditable Sandbox and Benchmark for Cross-User Agents Collaboration and Security in Human-Centered Agent Networks
Prince Zizhuang Wang, Aojie Yuan, Haiyue Zhang +3
Recent advances in persistent personal-agent frameworks are making human-centered agent networks realistic deployment targets: each user can be served by an AI agent that acts on t…
SEVA: Self-Evolving Verification Agent with Process Reward for Fact Attribution
Aojie Yuan, Yi Nian, Haiyue Zhang +2
Hallucination is the reliability bottleneck for LLM-based agents, and fact attribution verifiers are the last line of defense -- yet today's verifiers emit only opaque binary label…
AgentIR: A Workload-Adaptive Cascade Retrieval Substrate for Long-Term Conversational Memory
Aojie Yuan, Haiyue Zhang, Shahin Nazarian
Long-term conversational memory is a retrieval workload classical IR was not built for: the index grows during the query stream, query types shift intra-session, and the latency bu…
When Simulation Lies: A Sim-to-Real Benchmark and Domain-Randomized RL Recipe for Tool-Use Agents
Xiaolin Zhou, Aojie Yuan, Zheng Luo +12
Tool-use language agents are evaluated on benchmarks that assume clean inputs, unambiguous tool registries, and reliable APIs. Real deployments violate all these assumptions: user…
Hidden Error Awareness in Chain-of-Thought Reasoning: The Signal Is Diagnostic, Not Causal
Aojie Yuan, Zhiyuan Julian Su, Haiyue Zhang +2
Chain-of-thought (CoT) prompting assumes that generated reasoning reflects a model's internal computation. We show this assumption is wrong in a specific, measurable way: models in…
Auditable Agents
Yi Nian, Aojie Yuan, Haiyue Zhang +7
LLM agents call tools, query databases, delegate tasks, and trigger external side effects. Once an agent system can act in the world, the question is no longer only whether harmful…
Agent Audit: A Security Analysis System for LLM Agent Applications
Haiyue Zhang, Yi Nian, Yue Zhao
What should a developer inspect before deploying an LLM agent: the model, the tool code, the deployment configuration, or all three? In practice, many security failures in agent sy…
Sovereign-OS: A Charter-Governed Operating System for Autonomous AI Agents with Verifiable Fiscal Discipline
Aojie Yuan, Haiyue Zhang, Ziyi Wang +1
As AI agents evolve from text generators into autonomous economic actors that accept jobs, manage budgets, and delegate to sub-agents, the absence of runtime governance becomes a c…