all done

Author: unknown

paper/main.bib

@@ -758,6 +758,21 @@ @misc{Grok3_2025
 }
 
 
+@article{hu2025flowsearch,
+  title={FlowSearch: Advancing deep research with dynamic structured knowledge flow},
+  author={Hu, Yusong and Ma, Runmin and Fan, Yue and Shi, Jinxin and Cao, Zongsheng and Zhou, Yuhao and Yuan, Jiakang and Yan, Xiangchao and Zhang, Wenlong and Bai, Lei and others},
+  journal={arXiv preprint arXiv:2510.08521},
+  year={2025}
+}
+
+@article{shi2025dualresearch,
+  title={DualResearch: Entropy-Gated Dual-Graph Retrieval for Answer Reconstruction},
+  author={Shi, Jinxin and Cao, Zongsheng and Ma, Runmin and Hu, Yusong and Zhou, Jie and Li, Xin and Bai, Lei and He, Liang and Zhang, Bo},
+  journal={arXiv preprint arXiv:2510.08959},
+  year={2025}
+}
+
+
 @article{team2025novelseek,
   title={NovelSeek: When Agent Becomes the Scientist--Building Closed-Loop System from Hypothesis to Verification},
   author={Team, NovelSeek and Zhang, Bo and Feng, Shiyang and Yan, Xiangchao and Yuan, Jiakang and Yu, Zhiyin and He, Xiaohan and Huang, Songtao and Hou, Shaowei and Nie, Zheng and others},
@@ -800,4 +815,29 @@ @misc{zhao2025swiftascalablelightweightinfrastructure
       archivePrefix={arXiv},
       primaryClass={cs.CL},
       url={https://arxiv.org/abs/2408.05517}, 
+}
+
+@article{xu2025manalyzer,
+  title={Manalyzer: End-to-end Automated Meta-analysis with Multi-agent System},
+  author={Xu, Wanghan and Zhang, Wenlong and Ling, Fenghua and Fei, Ben and Hu, Yusong and Ren, Fangxuan and Lin, Jintai and Ouyang, Wanli and Bai, Lei},
+  journal={arXiv preprint arXiv:2505.20310},
+  year={2025}
+}
+
+@article{field2010meta,
+  title={How to do a meta-analysis},
+  author={Field, Andy P and Gillett, Raphael},
+  journal={British Journal of Mathematical and Statistical Psychology},
+  volume={63},
+  number={3},
+  pages={665--694},
+  year={2010},
+  publisher={Wiley Online Library}
+}
+
+@article{xu2025comprehensive,
+  title={A Comprehensive Survey of Deep Research: Systems, Methodologies, and Applications},
+  author={Xu, Renjun and Peng, Jingwen},
+  journal={arXiv preprint arXiv:2506.12594},
+  year={2025}
 }

paper/sections/0-abstract.tex

@@ -13,7 +13,7 @@
 
 Despite advances in scientific AI, a coherent framework for Scientific General Intelligence (SGI)—the ability to autonomously conceive, investigate, and reason across scientific domains—remains lacking. 
 We present a principled SGI definition grounded in the Practical Inquiry Model (PIM: Deliberation, Conception, Action, Perception) and operationalize it via four scientist-aligned tasks: deep research, idea generation, AI-assisted experiments (dry/wet), and experimental reasoning. 
-SGI-Bench comprises ~1{,}000 expert-curated, cross-disciplinary samples inspired by Science’s 125 Big Questions, enabling systematic evaluation of state-of-the-art LLMs. Results reveal gaps: low exact match (10--20\%) in deep research despite step-level alignment; ideas lacking feasibility and detail; high code executability but low execution result accuracy in dry experiments; low sequence fidelity in wet protocols; and persistent multimodal comparative-reasoning challenges. 
+SGI-Bench comprises over 1,000 expert-curated, cross-disciplinary samples inspired by Science’s 125 Big Questions, enabling systematic evaluation of state-of-the-art LLMs. Results reveal gaps: low exact match (10--20\%) in deep research despite step-level alignment; ideas lacking feasibility and detail; high code executability but low execution result accuracy in dry experiments; low sequence fidelity in wet protocols; and persistent multimodal comparative-reasoning challenges. 
 We further introduce Test-Time Reinforcement Learning (TTRL), which optimizes retrieval-augmented novelty rewards at inference, enhancing hypothesis novelty without reference answer. 
 Together, our PIM-grounded definition, workflow-centric benchmark, and empirical insights establish a foundation for AI systems that genuinely participate in scientific discovery.
 

paper/sections/1-introduction.tex

@@ -24,7 +24,7 @@ \section{Introduction}
 \item \textbf{Scientific Experimental Reasoning (Perception):} This task requires models to analyze experimental results, interpret data trends, and identify meaningful conclusions.
 \end{itemize}
 
-Building upon our theoretical framework, the construction of SGI-Bench operationalizes the proposed definition of \textbf{Scientific General Intelligence (SGI)}. We began with foundational topics drawn from \textit{Science’s 125 Big Questions for the 21st Century}~\cite{sanders2021125}, spanning ten major disciplinary areas. Through multi-round collaborations with domain experts, we identified high-impact, AI-assisted research problems and curated raw source materials from leading journals such as \textit{Nature}, \textit{Science}, and \textit{Cell}. Together with PhD-level researchers, we implemented a multi-stage quality control pipeline involving human annotation, model-based verification, and rule-based consistency checks. The resulting benchmark comprises approximately 1,000 expert-curated samples that concretely instantiate the reasoning, creativity, and experimental competencies central to our definition of SGI.
+Building upon our theoretical framework, the construction of SGI-Bench operationalizes the proposed definition of \textbf{Scientific General Intelligence (SGI)}. We began with foundational topics drawn from \textit{Science’s 125 Big Questions for the 21st Century}~\cite{sanders2021125}, spanning ten major disciplinary areas. Through multi-round collaborations with domain experts, we identified high-impact, AI-assisted research problems and curated raw source materials from leading journals such as \textit{Nature}, \textit{Science}, and \textit{Cell}. Together with PhD-level researchers, we implemented a multi-stage quality control pipeline involving human annotation, model-based verification, and rule-based consistency checks. The resulting benchmark comprises over 1,000 expert-curated samples that concretely instantiate the reasoning, creativity, and experimental competencies central to our definition of SGI.
 
 To evaluate performance across these four dimensions, we found that conventional “LLM-as-a-judge”~\cite{li2025generation} paradigms are insufficient to handle the diverse and specialized metrics required by SGI assessment. To address this, we developed an agent-based evaluation framework following an \textbf{Agent-as-a-judge}~\cite{zhuge2024agent} paradigm. Equipped with tools such as a web search interface, Python interpreter, file reader, PDF parser, and discipline-specific metric functions, this framework ensures rigor, scalability, and transparency. It operates through four interdependent stages—\textit{Question Selection}, \textit{Metric Customization}, \textit{Prediction \& Evaluation}, and \textit{Report Generation}—each coordinated by specialized agents aligned with different aspects of scientific inquiry.