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.
arXiv:2606. 19808v1 Announce Type: new Abstract: Test-time reasoning is increasingly used as a serving-time control knob, but extra reasoning is not uniformly valuable: it can repair failed attempts, waste compute on already-correct answers, or introduce harmful answer changes.
The paper introduces an action‑class diagnostic framework for multi‑turn tool‑calling in large language model agents, breaking failures into action‑class miscalibration and action‑execution failure across a four‑class action space (TOOL_CALL, ASK, REFUSE, CONFIRM). It defines a self‑revealing upper bound (Acc GAR) to expose state‑grader masking of miscalibration and shows that miscalibration is a significant, previously hidden failure mode, especially for heavily tool‑trained families. The study demonstrates that calibration can be reshaped by context‑only perturbations, but the effects vary widely across models and perturbation mechanisms, underscoring the need for diagnostics beyond aggregate accuracy.
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:2608. 05064v1 Announce Type: cross Abstract: Small open-weight language models increasingly run in private, offline, and cost-sensitive settings, where the key deployment question is not only what a model answers but when it should defer to a human.
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 "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:2609. 29140v1 Announce Type: new Abstract: Repeated evaluation can estimate a benchmark score accurately while still requiring replication to certify narrow uncertainty.
The paper examines three recent studies that highlight distinct reliability issues in large language models (LLMs). Each study points to a missing capability—whether a tool‑reliability representation, safe generation behavior, or a consistent‑reasoning function—yet all converge on the need for calibrated abstention. The authors argue that current benchmarks fail to reward abstention, preventing the development of this function, and propose four evaluation changes to address the gap.
arXiv:2607. 25398v1 Announce Type: new Abstract: Language-model agents are increasingly deployed under standing instructions: a system prompt, a policy file, or a skills document is placed in context, and the agent is trusted to let it govern every action that follows.
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...
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. 23386v1 Announce Type: new Abstract: We document a failure class in frontier large language models -- exception chain collapse -- observed in eligibility evaluation under nested conditional rules of the form "A is required UNLESS B applies, UNLESS C overrides B".