← Back to all news
arXiv AI

Same Patient, Different Order: Action-Level Reliability of Clinical LLM Agents Under Repeated Runs

The paper introduces a new evaluation method called "same-input rerun" to assess the consistency of clinical language‑model agents across repeated runs. By replaying 1,000 MedAgentBench tasks with identical inputs, the authors find that action‑level outputs—such as test orders, medication requests, and referrals—vary significantly, even when benchmark scores remain unchanged. The study demonstrates that current benchmarks, which typically evaluate only a single run per task, can miss substantial behavioral divergence.

Related stories

arXiv Machine Learning
Sep 30

Frozen Judges, Moving Agents: Version-Dependent LLM-Judge Error and the Limits of Judge-Assisted Agent Evaluation

The paper investigates how language‑model judges can make version‑dependent errors when evaluating upgraded agents. Using 35 public coding‑agent submissions, two customer‑service agents, and over a thousand expert‑labeled trajectories, the authors show that fixed judges often reject task‑conditioned error invariance and can incorrectly approve failed patches, especially as agent capability increases. Paired audits of current outputs reduce interval width only marginally, and the study concludes that independent human patch review is still necessary.

By Jiapeng Li
llmsagents
More like this →
arXiv AI
Aug 28

Same Model, Different Harness: Different Coding-Agent Results

The paper investigates how altering the harness—specifically the way a coding agent manages context and tool outputs—affects performance when the underlying model and task remain unchanged. Two harness configurations were compared on three coding benchmarks: a control that preserves the full conversation in order, and a treatment that mechanically shortens older tool results to keep the context tight. Across all benchmarks, the treatment increased the mean per‑task fail‑to‑pass fraction and, in some cases, the number of complete solutions, demonstrating that the harness itself can significantly influence a frozen model’s effectiveness.

By Sydney Lewis
agentsbenchmarks
More like this →
arXiv Computation and Language
Oct 9

TRACE: Diagnosing Verifier Brittleness in Agentic Evaluation

The paper introduces TRACE, a protocol designed to diagnose whether changes in verifier scores for large language model agents reflect actual changes in agent behavior or merely alterations in the evaluation process. TRACE works by applying targeted changes to evaluation components, running paired experiments, and re‑scoring unchanged trajectories to isolate the source of score variation. Experiments on synthetic tasks and public benchmark tasks demonstrate that seemingly significant score shifts can often be attributed to evaluation artifacts, while TRACE can also detect genuine behavioral changes.

By Radhika Gaonkar
llmsagentsbenchmarks
More like this →
arXiv AI
Oct 16

Auditing Escalation Reliability in Healthcare LLM Agents: A Perturbation-Controlled Benchmark of Workflow-State Failures

The paper introduces a benchmark of 400 synthetic healthcare workflow episodes to evaluate large language model (LLM) agents’ reliability in escalating care when evidence is missing or conflicting. It finds that while clean decision accuracy is high (0.97), models miss a large proportion of necessary escalations (32–97%) under uncertainty, and even the largest 9.7B model’s performance drops to 0.78 in conflict scenarios. A workflow-aware controller improves utility and reduces missed escalations, but its effectiveness is limited by lexical generalization failures.

By Zihao Zheng (Washington University in St. Louis), Junyi Yao (Washington University in St. Louis), Baichuan Li (Southern Methodist University), Jiayu Long (Washington University in St. Louis)
llmsagentsbenchmarks
More like this →
arXiv AI
Oct 26

Async-RBench: Benchmarking Asynchronous Task Execution in LLM Agents

Async-RBench is a new benchmark that tests how large language model agents handle asynchronous task execution. It includes 200 executable tasks across eight scenarios, with a scenario controller that delivers results based on task-state conditions and an Event Response Contract that defines required changes, preservation, avoidance, and re-verification. The benchmark introduces the Dynamic Replanning Score (DRS) to evaluate both the response process and outcome, and finds that agents generally perform worse with asynchronous delivery compared to batched delivery, highlighting challenges in reliable asynchronous task handling.

By Detao Zeng, Guilin Ou, Nuo Chen, Yunhao Feng, Hongliang Wang, Qinli Yang, Junming Shao, Zhong Zhang
llmsagentsbenchmarks
More like this →
arXiv AI
Aug 28

AgentJudgeBench: A Multi-Difficulty Benchmark for Evaluating LLM Judges on Agentic Tool-Calling

AgentJudgeBench is a new benchmark that evaluates the reliability of large language model (LLM) judges on agentic tool‑calling tasks involving workflow directed acyclic graphs (DAGs). It contains 3,808 instances across six DAG topologies and three difficulty tiers, tested with five generators (3B–70B open‑weight models and GPT‑5.4) and six judges (20B to frontier scale) under both paired‑with‑and‑without‑ground‑truth conditions. The study finds that judge alignment degrades with task difficulty, ground‑truth exposure can sometimes hurt alignment, and structured evaluation rubrics provide modest improvements, revealing a structural ceiling that model capacity alone cannot surpass.

By Abhigya Verma, Amit Kumar Saha, Seganrasan Subramanian, Sai Harshitha Aluru
llmsagentsbenchmarkssafety
More like this →