LLM agents are increasingly evaluated on multi-week decision tasks in which the state that drives cost is never directly observed. On such tasks the final cost cannot say why an agent failed: it may have misread the world, or read it correctly and still failed to act (the knowing-doing gap).
arXiv:2607. 13618v1 Announce Type: new Abstract: LLM agents are increasingly evaluated on multi-week decision tasks in which the state that drives cost is never directly observed.
arXiv:2606. 29654v1 Announce Type: new Abstract: Multi-agent deliberation among LLMs can improve reasoning, but deployment requires deciding when the current answer is reliable enough to act on and when it should be escalated to human review.
The paper "trajectory-judge: What Outcome-Only LLM Judges Miss on Agent Trajectories" examines the limitations of outcome-only evaluation for large language model agents. Using a deterministic tool‑using support‑desk environment with a scripted oracle policy and a fault injector, the authors compare five different judging approaches—programmatic rules, outcome‑only, step‑rubric at two model sizes, and a self‑consistency ensemble—on metrics such as detection, step localisation, fault typing, calibration, and cost across 400 trajectories. The study finds that outcome‑only judges miss many silent faults and generate false positives, while step‑rubric judges achieve higher recall with no false alarms but at greater cost, and that none of the judges read the final reply, allowing fabricated promises to evade detection. "whyItMatters":"The findings highlight that current production‑default outcome‑only evaluations can overlook critical failures in agent behavior, underscoring the need for more nuanced, step‑level judging methods to ensure reliable LLM agent performance."
arXiv:2607. 17044v1 Announce Type: cross Abstract: Multi-step enterprise agent tasks fail in a characteristic way: single-pass inference has no checkpoint between deciding an answer and committing to it.
The paper investigates the reliability of tool‑using agents, focusing on two failure modes: selecting the wrong tool and constructing incorrect arguments. It introduces a correct‑invocation rate metric to distinguish these errors and evaluates five open‑weight models on multi‑step tasks up to depth 8, finding that by depth 6 about 70% of a model’s clean‑context capability is lost due to earlier mistakes. The study reveals that exact‑match scoring against a fixed gold trajectory forces severity and recovery parameters to extreme values, and proposes a conditional‑on‑state scoring remedy that yields more realistic severity estimates.
arXiv:2607. 04419v1 Announce Type: new Abstract: Most agent evaluations collapse a multi-step trace into a final answer, a success flag, or a trajectory-level score.
arXiv:2609.25686v1 Announce Type: cross Abstract: Long-horizon assigned work requires an LLM agent to track the state of a task: which steps are done, blocked, cancelled, or open to repetition. Agent...
arXiv:2608. 15286v1 Announce Type: cross Abstract: We introduce AgentRelBench, an environment-agnostic reliability instrument that computes ground-truth, severity-priced damage from database state diffs across repeated runs, with no LLM in the measurement path, demonstrated on EnterpriseOps-Gym.
arXiv:2609.05435v1 Announce Type: new Abstract: Modern language agents are expected to operate over long horizons: they ask follow-up questions, reuse worked examples, handle tool feedback, and adapt...
arXiv:2607. 17136v1 Announce Type: cross Abstract: Agentic computer-use RL is reported in single runs, and those numbers mislead.
The paper proposes a claim‑specific verification audit for modular agents that replaces aggregate task scores with evidence‑based evaluations. Each agent conclusion is recorded with supporting evidence and classified as supported, unsupported, unresolved, or not evaluated, along with the boundary of validity. The audit employs three tools—oracle policies, perfect component replacements, and verifier‑score tests—to trace value changes, locate lost value, and assess verifier effectiveness, demonstrated on a portfolio‑allocation agent in a synthetic market.
A 0.6B language model consistently answers YES to 1,200 logical tests, yet its behavior shows no discrimination. Linear probes reveal the correct verdict with high AUC (0.96) and transfer to unseen structures, but a single scalar readout fails due to a saturated decision threshold offset by +4.6 σ. Adjusting this threshold restores behavior accuracy from 50 % to 81 % and improves higher‑scale models, demonstrating that miscalibrated readouts, not hidden knowledge loss, drive performance gaps.