mlsanityData to 14 Oct 2026

Perspectra: Choosing Your Experts Enhances Critical Thinking in Multi-Agent Research Ideation

arXiv 2509.20553 ↗First posted 24 Sep 2025cs.HC cs.AI cs.CLcited by 4

On the map: LLM reasoning → Reasoning for agents, planning and code

also in 1 more
Abstract (arXiv)

Recent advances in multi-agent systems (MAS) enable tools for information search and ideation by assigning personas to agents. However, how users can effectively control, steer, and critically evaluate collaboration among multiple domain-expert agents remains underexplored. We present Perspectra, an interactive MAS that visualizes and structures deliberation among LLM agents via a forum-style interface, supporting @-mention to invite targeted agents, threading for parallel exploration, with a real-time mind map for visualizing arguments and rationales. In a within-subjects study with 18 participants, we compared Perspectra to a group-chat baseline as they developed research proposals. Our findings show that Perspectra significantly increased the frequency and depth of critical-thinking behaviors, elicited more interdisciplinary replies, and led to more frequent proposal revisions than the group chat condition. We discuss implications for designing multi-agent tools that scaffold critical thinking by supporting user control over multi-agent adversarial discourse.

Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Placed in Reasoning for agents, planning and code, one of 17 directions in LLM reasoning. Builds on 30 of its 93 references that are in the map. First posted on arXiv 24 Sep 2025.

Builds on

30 of 93 references.

Closest ancestors · 3

Same direction, newest first

  1. 2025MetaCrit: A Critical Thinking Framework for Self-Regulated LLM Reasoning20 Jul
  2. 2025Conversational Agents as Catalysts for Critical Thinking: Challenging Social Influence in Group Decision-making18 Mar
  3. 2024SocraSynth: Multi-LLM Reasoning with Conditional Statistics19 Jan

Other references · 27

  1. 2023Towards Understanding Sycophancy in Language Modelscited by 1,040
  2. 2024Large Language Model based Multi-Agents: A Survey of Progress and Challengescited by 769
  3. 2024Scaling Synthetic Data Creation with 1,000,000,000 Personascited by 417
  4. 2024LightRAG: Simple and Fast Retrieval-Augmented Generationcited by 315
  5. 2025DeepResearcher: Scaling Deep Research via Reinforcement Learning in Real-world Environmentscited by 286
  6. 2024Magentic-One: A Generalist Multi-Agent System for Solving Complex Taskscited by 232
  7. 2024ResearchAgent: Iterative Research Idea Generation over Scientific Literature with Large Language Modelscited by 210
  8. 2025Which Agent Causes Task Failures and When? On Automated Failure Attribution of LLM Multi-Agent Systemscited by 185
  9. 2022OpenAlex: A fully-open index of scholarly works, authors, venues, institutions, and conceptscited by 177
  10. 2025Deep Research Agents: A Systematic Examination And Roadmapcited by 128
  11. 2024Simulating Classroom Education with LLM-Empowered Agentscited by 127
  12. 2025A Survey of Context Engineering for Large Language Modelscited by 126
15 more
  1. 2025G-Memory: Tracing Hierarchical Memory for Multi-Agent Systemscited by 122
  2. 2023The Semantic Scholar Open Data Platformcited by 118
  3. 2025InternAgent: When Agent Becomes the Scientist -- Building Closed-Loop System from Hypothesis to Verificationcited by 46
  4. 2024Into the Unknown Unknowns: Engaged Human Learning through Participation in Language Model Agent Conversationscited by 43
  5. 2024Human-LLM Compound System for Scientific Ideation through Facet Recombination and Novelty Evaluationcited by 41
  6. 2025MDTeamGPT: A Self-Evolving LLM-based Multi-Agent Framework for Multi-Disciplinary Team Medical Consultationcited by 26
  7. 2025IRIS: Interactive Research Ideation System for Accelerating Scientific Discoverycited by 19
  8. 2024Argumentative Experience: Reducing Confirmation Bias on Controversial Issues through LLM-Generated Multi-Persona Debatescited by 15
  9. 2025DBox: Scaffolding Algorithmic Programming Learning through Learner-LLM Co-Decompositioncited by 9
  10. 2025From Conversation to Orchestration: HCI Challenges and Opportunities in Interactive Multi-Agentic Systemscited by 9
  11. 2024Conversational Agents as Catalysts for Critical Thinking: Challenging Design Fixation in Group Designcited by 4
  12. 2025From Consumption to Collaboration: Measuring Interaction Patterns to Augment Human Cognition in Open-Ended Taskscited by 3
  13. 2025Large Language Model-Empowered Interactive Load Forecastingcited by 2
  14. 2025Scaffolding Recursive Divergence and Convergence in Story Ideationcited by 2
  15. 2023Fostering User Engagement in the Critical Reflection of Argumentscited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 1; 2026: 3.

4 papers.

  1. 2025ScholarEval: Research Idea Evaluation Grounded in Literaturecited by 12
  2. 2026AgentPanel: Toward a New Paradigm for Human--AI Collaboration in Exploring Scientific Questionscited by 1
  3. 2026Designing Proactive Thought Partners for Writingcited by 1
  4. 2026DocuTeam: Mixed-Initiative Multi-Agent Discussions around Evolving Documentscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Exploring Design of Multi-Agent LLM Dialogues for Research Ideation
  2. 2025Thinking with Many Minds: Using Large Language Models for Multi-Perspective Problem-Solving
  3. 2026The Virtual Roundtable: Multi-Agent Personas Simulating the Dynamics of Human Brainstorming
  4. 2025Revisiting Multi-Agent Debate as Test-Time Scaling: A Systematic Study of Conditional Effectiveness
  5. 2026From Diagnosis to Redesign: Using Quantitative Ethnography to Improve Multi-Agent LLM Reasoning
  6. 2025Can LLM Agents Really Debate? A Controlled Study of Multi-Agent Debate in Logical Reasoning
  7. 2025Enhancing LLM Performance Through Debate: An Empirical Study on Multi-Agent Debate for Coding Tasks
  8. 2026Ideation Arena: Evaluating LLM Generated Research Ideas with Battle-style Human Expert Assessment