SciSage: A Multi-Agent Framework for High-Quality Scientific Survey Generation

• URL: http://arxiv.org/abs/2506.12689v2
• Date: Mon, 21 Jul 2025 03:49:38 GMT
• Title: SciSage: A Multi-Agent Framework for High-Quality Scientific Survey Generation
• Authors: Xiaofeng Shi, Qian Kou, Yuduo Li, Ning Tang, Jinxin Xie, Longbin Yu, Songjing Wang, Hua Zhou,
• Abstract summary: SciSage is a multi-agent framework employing a reflect-when-you-write paradigm.<n>It critically evaluates drafts at outline, section, and document levels, collaborating with specialized agents for query interpretation, content retrieval, and refinement.<n>We also release SurveyScope, a benchmark of 46 high-impact papers ( 2020-2025) across 11 computer science domains.
• Score: 2.985620880452744
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The rapid growth of scientific literature demands robust tools for automated survey-generation. However, current large language model (LLM)-based methods often lack in-depth analysis, structural coherence, and reliable citations. To address these limitations, we introduce SciSage, a multi-agent framework employing a reflect-when-you-write paradigm. SciSage features a hierarchical Reflector agent that critically evaluates drafts at outline, section, and document levels, collaborating with specialized agents for query interpretation, content retrieval, and refinement. We also release SurveyScope, a rigorously curated benchmark of 46 high-impact papers (2020-2025) across 11 computer science domains, with strict recency and citation-based quality controls. Evaluations demonstrate that SciSage outperforms state-of-the-art baselines (LLM x MapReduce-V2, AutoSurvey), achieving +1.73 points in document coherence and +32% in citation F1 scores. Human evaluations reveal mixed outcomes (3 wins vs. 7 losses against human-written surveys), but highlight SciSage's strengths in topical breadth and retrieval efficiency. Overall, SciSage offers a promising foundation for research-assistive writing tools.

Related papers

• ScholarPeer: A Context-Aware Multi-Agent Framework for Automated Peer Review [48.60540055009675]
  ScholarPeer is a search-enabled multi-agent framework designed to emulate the cognitive processes of a senior researcher.<n>We evaluate ScholarPeer on DeepReview-13K and the results demonstrate that ScholarPeer achieves significant win-rates against state-of-the-art approaches in side-by-side evaluations.
  arXiv  Detail & Related papers  (2026-01-30T06:54:55Z)
• Can Deep Research Agents Find and Organize? Evaluating the Synthesis Gap with Expert Taxonomies [57.11324429385405]
  We introduce TaxoBench, a diagnostic benchmark derived from 72 computer science surveys.<n>We manually extract expert-authored taxonomy trees containing 3,815 precisely categorized citations as ground truth.<n>Best agent recalls only 20.9% of expert-selected papers, and even with perfect input, the best model achieves only 0.31 ARI in organization.
  arXiv  Detail & Related papers  (2026-01-18T11:57:09Z)
• RPC-Bench: A Fine-grained Benchmark for Research Paper Comprehension [65.81339691942757]
  RPC-Bench is a large-scale question-answering benchmark built from review-rebuttal exchanges of high-quality computer science papers.<n>We design a fine-grained taxonomy aligned with the scientific research flow to assess models' ability to understand and answer why, what, and how questions in scholarly contexts.
  arXiv  Detail & Related papers  (2026-01-14T11:37:00Z)
• SciFig: Towards Automating Scientific Figure Generation [41.73701976318102]
  SciFig is an end-to-end AI agent system that generates publication-ready pipeline figures directly from research paper texts.<n>We introduce a rubric-based evaluation framework that analyzes 2,219 real scientific figures to extract evaluation rubrics.<n>SciFig demonstrates remarkable performance: achieving 70.1$%$ overall quality on dataset-level evaluation and 66.2$%$ on paper-specific evaluation.
  arXiv  Detail & Related papers  (2026-01-07T20:56:58Z)
• OpenNovelty: An LLM-powered Agentic System for Verifiable Scholarly Novelty Assessment [63.662126457336534]
  OpenNovelty is an agentic system for transparent, evidence-based novelty analysis.<n>It grounds all assessments in retrieved real papers, ensuring verifiable judgments.<n>OpenNovelty aims to empower the research community with a scalable tool that promotes fair, consistent, and evidence-backed peer review.
  arXiv  Detail & Related papers  (2026-01-04T15:48:51Z)
• SurveyBench: Can LLM(-Agents) Write Academic Surveys that Align with Reader Needs? [37.28508850738341]
  Survey writing is a labor-intensive and intellectually demanding task.<n>Recent approaches, such as general DeepResearch agents and survey-specialized methods, can generate surveys automatically.<n>But their outputs often fall short of human standards and there lacks a rigorous, reader-aligned benchmark.<n>We propose a fine-grained, quiz-driven evaluation framework SurveyBench.
  arXiv  Detail & Related papers  (2025-10-03T15:49:09Z)
• LiRA: A Multi-Agent Framework for Reliable and Readable Literature Review Generation [66.09346158850308]
  We present LiRA (Literature Review Agents), a multi-agent collaborative workflow which emulates the human literature review process.<n>LiRA utilizes specialized agents for content outlining, subsection writing, editing, and reviewing, producing cohesive and comprehensive review articles.<n>We evaluate LiRA in real-world scenarios using document retrieval and assess its robustness to reviewer model variation.
  arXiv  Detail & Related papers  (2025-10-01T12:14:28Z)
• Agentic AutoSurvey: Let LLMs Survey LLMs [31.13281166545961]
  We present textbfAgentic AutoSurvey, a multi-agent framework for automated survey generation.<n>Our system employs four specialized agents (Paper Search Specialist, Topic Mining & Clustering, Academic Survey Writer, and Quality Evaluator) working in concert to generate comprehensive literature surveys with superior synthesis quality.
  arXiv  Detail & Related papers  (2025-09-23T05:28:43Z)
• SurveyGen: Quality-Aware Scientific Survey Generation with Large Language Models [14.855783196702191]
  We present SurveyGen, a large-scale dataset comprising over 4,200 human-written surveys across diverse scientific domains.<n>We build QUAL-SG, a novel quality-aware framework for survey generation.
  arXiv  Detail & Related papers  (2025-08-25T04:22:23Z)
• SurGE: A Benchmark and Evaluation Framework for Scientific Survey Generation [37.921524136479825]
  SurGE (Survey Generation Evaluation) is a new benchmark for scientific survey generation in computer science.<n>SurGE consists of (1) a collection of test instances, each including a topic description, an expert-written survey, and its full set of cited references, and (2) a large-scale academic corpus of over one million papers.<n>In addition, we propose an automated evaluation framework that measures the quality of generated surveys across four dimensions.
  arXiv  Detail & Related papers  (2025-08-21T15:45:10Z)
• Automated Thematic Analyses Using LLMs: Xylazine Wound Management Social Media Chatter Use Case [2.583403860629219]
  Large language models (LLMs) face challenges in inductive thematic analysis.<n>We evaluated the feasibility of using LLMs to replicate expert-driven thematic analysis of social media data.
  arXiv  Detail & Related papers  (2025-07-14T20:57:52Z)
• SciArena: An Open Evaluation Platform for Foundation Models in Scientific Literature Tasks [87.29946641069068]
  We present SciArena, an open and collaborative platform for evaluating foundation models on scientific literature tasks.<n>By leveraging collective intelligence, SciArena offers a community-driven evaluation of model performance on open-ended scientific tasks.<n>We release SciArena-Eval, a meta-evaluation benchmark based on our collected preference data.
  arXiv  Detail & Related papers  (2025-07-01T17:51:59Z)
• LLM-Based Compact Reranking with Document Features for Scientific Retrieval [30.341167520613197]
  We propose a training-free, model-agnostic semantic reranking framework for scientific retrieval called CoRank.<n>CoRank involves three stages: offline extraction of document-level features, coarse reranking using these compact representations, and fine-grained reranking on full texts of the top candidates from stage.<n> Experiments on LitSearch and CSFCube show that CoRank significantly improves reranking performance across different LLM backbones.
  arXiv  Detail & Related papers  (2025-05-19T22:10:27Z)
• XtraGPT: LLMs for Human-AI Collaboration on Controllable Academic Paper Revision [41.44785777328187]
  XtraGPT is the first suite of open-source large language models (LLMs) designed to provide context-aware, instruction-guided writing assistance.<n>We introduce a dataset of 7,040 research papers from top-tier venues annotated with over 140,000 instruction-response pairs.<n>Experiments validate that XtraGPT significantly outperforms same-scale baselines and approaches the quality of proprietary systems.
  arXiv  Detail & Related papers  (2025-05-16T15:02:19Z)
• SurveyForge: On the Outline Heuristics, Memory-Driven Generation, and Multi-dimensional Evaluation for Automated Survey Writing [13.101632066188532]
  We introduce SurveyForge, which generates the outline by analyzing the logical structure of human-written outlines.<n>To achieve a comprehensive evaluation, we construct SurveyBench, which includes 100 human-written survey papers for win-rate comparison.<n>Experiments demonstrate that SurveyForge can outperform previous works such as AutoSurvey.
  arXiv  Detail & Related papers  (2025-03-06T17:15:48Z)
• TrustRAG: An Information Assistant with Retrieval Augmented Generation [73.84864898280719]
  TrustRAG is a novel framework that enhances acRAG from three perspectives: indexing, retrieval, and generation.<n>We open-source the TrustRAG framework and provide a demonstration studio designed for excerpt-based question answering tasks.
  arXiv  Detail & Related papers  (2025-02-19T13:45:27Z)
• Automating Bibliometric Analysis with Sentence Transformers and Retrieval-Augmented Generation (RAG): A Pilot Study in Semantic and Contextual Search for Customized Literature Characterization for High-Impact Urban Research [2.1728621449144763]
  Bibliometric analysis is essential for understanding research trends, scope, and impact in urban science. Traditional methods, relying on keyword searches, often fail to uncover valuable insights not explicitly stated in article titles or keywords. We leverage Generative AI models, specifically transformers and Retrieval-Augmented Generation (RAG), to automate and enhance bibliometric analysis.
  arXiv  Detail & Related papers  (2024-10-08T05:13:27Z)
• Are Large Language Models Good Classifiers? A Study on Edit Intent Classification in Scientific Document Revisions [62.12545440385489]
  Large language models (LLMs) have brought substantial advancements in text generation, but their potential for enhancing classification tasks remains underexplored. We propose a framework for thoroughly investigating fine-tuning LLMs for classification, including both generation- and encoding-based approaches. We instantiate this framework in edit intent classification (EIC), a challenging and underexplored classification task.
  arXiv  Detail & Related papers  (2024-10-02T20:48:28Z)
• ResearchArena: Benchmarking Large Language Models' Ability to Collect and Organize Information as Research Agents [21.17856299966841]
  This study introduces ResearchArena, a benchmark designed to evaluate large language models (LLMs) in conducting academic surveys.<n>To support these opportunities, we construct an environment of 12M full-text academic papers and 7.9K survey papers.
  arXiv  Detail & Related papers  (2024-06-13T03:26:30Z)
• SciRIFF: A Resource to Enhance Language Model Instruction-Following over Scientific Literature [80.49349719239584]
  We present SciRIFF (Scientific Resource for Instruction-Following and Finetuning), a dataset of 137K instruction-following demonstrations for 54 tasks. SciRIFF is the first dataset focused on extracting and synthesizing information from research literature across a wide range of scientific fields.
  arXiv  Detail & Related papers  (2024-06-10T21:22:08Z)
• A Literature Review of Literature Reviews in Pattern Analysis and Machine Intelligence [55.33653554387953]
  Pattern Analysis and Machine Intelligence (PAMI) has led to numerous literature reviews aimed at collecting and fragmented information.<n>This paper presents a thorough analysis of these literature reviews within the PAMI field.<n>We try to address three core research questions: (1) What are the prevalent structural and statistical characteristics of PAMI literature reviews; (2) What strategies can researchers employ to efficiently navigate the growing corpus of reviews; and (3) What are the advantages and limitations of AI-generated reviews compared to human-authored ones.
  arXiv  Detail & Related papers  (2024-02-20T11:28:50Z)
• PROXYQA: An Alternative Framework for Evaluating Long-Form Text Generation with Large Language Models [72.57329554067195]
  ProxyQA is an innovative framework dedicated to assessing longtext generation. It comprises in-depth human-curated meta-questions spanning various domains, each accompanied by specific proxy-questions with pre-annotated answers. It assesses the generated content's quality through the evaluator's accuracy in addressing the proxy-questions.
  arXiv  Detail & Related papers  (2024-01-26T18:12:25Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.