一作ACL 2026Poster· Natural Language Processing
Beyond Single-shot Writing: Deep Research Agents are Unreliable at Multi-turn Report Revision
Bingsen Chen · Boyan Li · Ping Nie · Yuyu Zhang · Xi Ye · Chen Zhao
摘要
Existing benchmarks for Deep Research Agents (DRAs) treat report generation as a single-shot writing task, which fundamentally diverges from how human researchers iteratively draft and revise reports via self-reflection or peer feedback. Whether DRAs can reliably revise reports with user feedback...