{
  "id": 10615513,
  "title": "Failure-Transparent Agents: Benchmarking Post-Failure Reporting in Tool-Using Language Models",
  "url": "https://urgent.news/2026/09/28/failure-transparent-agents-benchmarking-post-failure-reporting-in",
  "topic": "ai",
  "section": "AI",
  "published": "2026-09-28T17:51:41.000Z",
  "source": {
    "name": "arXiv cs.AI",
    "slug": "arxiv-cs-ai",
    "url": "https://arxiv.org/abs/2609.35732v1"
  },
  "original_language": "en",
  "account": null,
  "summary": "Tool-using agents can fail twice: a required tool can fail, and the agent can then report success without the evidence needed to justify it. Existing benchmarks often entangle this reporting failure with tool selection, recovery, and environment dynamics. We introduce Failure-Transparent Agents (FTA), a controlled benchmark that fixes the failed observation and required evidence state before…",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}