Beyond Single-shot Writing: Deep Research Agents are Unreliable at Multi-turn Report Revision
cs.CL, cs.AI
Submitted: 2026-01-19
Updated: 2026-11-02
Code: https://github.com/BaleChen/Mr-Dre
Terminology
Sources
- HealthBench: Evaluating Large Language Models Towards Improved Human Health
- BrowseComp-Plus: A More Fair and Transparent Evaluation Benchmark of Deep-Research Agent
- Understanding DeepResearch via Reports
- CRITIC: Large Language Models Can Self-Correct with Tool-Interactive Critiquing
- Deep Researcher with Test-Time Diffusion
- Large Language Models Cannot Self-Correct Reasoning Yet
- ReportBench: Evaluating Deep Research Agents via Academic Survey Tasks
- WebThinker: Empowering Large Reasoning Models with Deep Research Capability
- GAIA: a benchmark for General AI Assistants
- Humanity's Last Exam
- WebResearcher: Unleashing unbounded reasoning capability in Long-Horizon Agents
- ExpertLongBench: Benchmarking Language Models on Expert-Level Long-Form Generation Tasks with Structured Checklists
- DR Tulu: Reinforcement Learning with Evolving Rubrics for Deep Research
- ResearchRubrics: A Benchmark of Prompts and Rubrics For Evaluating Deep Research Agents
- Tongyi DeepResearch Technical Report
- LiveResearchBench: A Live Benchmark for User-Centric Deep Research in the Wild
- BrowseComp: A Simple Yet Challenging Benchmark for Browsing Agents
- ResearcherBench: Evaluating Deep AI Research Systems on the Frontiers of Scientific Inquiry
- Dr. Bench: A Multidimensional Evaluation for Deep Research Agents, from Answers to Reports
- Instruction-Following Evaluation for Large Language Models
Related papers
- ALARM: Audio-Language Alignment for Reasoning Models
- ThinkSum: Probabilistic reasoning over sets using large language models
- Coherence boosting: When your pretrained language model is not paying enough attention
- From Facts to Insights: A Persona-Driven Dual Memory Framework and Dataset for Role-Playing Agents
- Improving General Role-Playing Agents via Psychology-Grounded Reasoning and Role-Aware Policy Optimization
- The Art of Asking: Multilingual Prompt Optimization for Synthetic Data