{
  "id": 3635,
  "url": "https://arxiv.org/abs/2605.27820v1",
  "title": "EgoBench: An Interactive Egocentric Multimodal Benchmark for Tool-Using Agents",
  "summary": "As AI agents increasingly operate in open, real-world environments, they require a deep synergy of multimodal perception, tool invocation with multi-hop reasoning, and dynamic interaction with users. However, existing benchmarks fail to jointly evaluate these capabilities due to challenges in designing strictly coupled multi-capability tasks, simulating natural and task-constrained user feedback, and ensuring objective evaluation of dynamic interaction. To bridge this gap, we introduce EgoBench,",
  "authors": "Yunqi Liu, Tong Niu, Zitong Wang, Zhenlong Dai, Yuqi Qing, Weiqiang Wang et al.",
  "category": "research",
  "topics": "agents-autonomy,environment",
  "orgs": null,
  "regions": null,
  "published_at": "2026-05-27T01:28:15.000Z",
  "fetched_at": "2026-07-14T16:30:23.247Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/3635",
  "original_url": "https://arxiv.org/abs/2605.27820v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}