RubricAtlas is a framework that extracts type-level analytical rubrics from expert reports to evaluate deep research agents. It organizes binary checkpoints around recurring analytical questions, linking them to report passages and assessing completeness separately from argument validity. Experiments on 3,079 Chinese financial reports show that individual reports cover only about half of their type’s questions, and that using different reports for requirements changes evaluation scores and agent rankings.
By Delin Wang, Sinuo Wang, Rui Sun, Beidi Luan, Qiming Shi, Jing Li, Daxin Jiang, Zuo Bai
Deep Research Bench II is a new benchmark designed to evaluate Deep Research Agents (DRAs) by requiring them to produce research reports for 132 grounded tasks across 22 domains. Each report is assessed using 9,430 fine‑grained binary rubrics that cover information recall, analysis, and presentation, all derived from expert‑written investigative articles through a rigorous LLM‑plus‑human pipeline. Evaluation of current state‑of‑the‑art DRAs shows that even the best models satisfy fewer than 50% of these rubrics, highlighting a significant gap between automated agents and human experts.
By Ruizhe Li, Mingxuan Du, Benfeng Xu, Chiwei Zhu, Xiaorui Wang, Zhendong Mao
arXiv:2610.18092v1 Announce Type: new
Abstract: Deep research (DR) agents address open-ended information needs through multi-step reasoning, long-horizon web search, and evidence synthesis. However,...
By Yibo Wang, Senjie Jin, Yansong Ning, Xinbing Liang, Haotian Luo, Jiazheng Zhang, Yuxuan He, Will Wu, Clarenceai
FinAutoRubric is a system that lets experts define reusable evaluation guidance for financial research agents, which the agents then use to automatically generate, review, and validate task‑specific rubrics. The approach replaces costly fixed rubrics with query‑specific ones that incorporate expert standards and can be updated as new information becomes available. On three expert‑authored finance benchmarks, the automatically generated rubrics closely match human scoring and are preferred by in‑house analysts in blind reviews.
By Hoyoung Lee, Suyeol Yun, Jack Haverty, Yunju Cho, Meesong Kim, Daekyung Park, Sumin Kim, Jihoon Kwon, Jasmine Jia Geng, Andrew Chin, Yin Luo, Edward Tong, Yu Yu, Zach Golkhou, Minkyu Kim, Igor Halperin, Young Cha, Alejandro Lopez-Lira, Chanyeol Choi, Yongjae Lee
AgentHop is a diagnostic benchmark comprising 1,011 multiple‑choice questions that test agentic multi‑hop question answering over scientific citation graphs. It provides a controlled seven‑tool sandbox with fixed token, turn, and tool‑call limits, allowing researchers to dissect model performance across four operational axes: retrieval, synthesis, tool‑call, and resource management. Analysis of 19 models shows distinct family‑specific behaviors, such as GPT models committing early, Anthropic and GLM checkpoints verifying before committing, DeepSeek and Kimi over‑searching, and Gemini‑3 Pro maintaining balance, while within‑family differences are revealed in retrieval versus synthesis strategies.
By Chanhee Park, Jeongho Yoon, Sungbin Han, Hyeonseok Moon, Heuiseok Lim
arXiv:2606. 08671v1 Announce Type: new Abstract: Agent skills extend language-model agents with task-specific procedures, scripts, and references, but the tasks and environments they target continually change.
By Zhiwei Li, Yong Hu