mlsanityData to 14 Oct 2026

MetaCrit: A Critical Thinking Framework for Self-Regulated LLM Reasoning

arXiv 2507.15015 ↗First posted 20 Jul 2025cs.MAcited by 1

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

also in 1 more
Abstract (arXiv)

Large language models (LLMs) fail on over one-third of multi-hop questions with counterfactual premises and remain vulnerable to adversarial prompts that trigger biased or factually incorrect responses, which exposes a fundamental deficit in self-regulated reasoning. We propose MetaCrit, a multi-agent framework grounded in Nelson and Narens' metacognitive regulation theory. MetaCrit decomposes reasoning regulation into four agents: object-level generation, a monitoring agent that assesses response validity, a control agent that critiques logical soundness, and a meta-level synthesizer that integrates all signals into a final response. Evaluation across eight benchmarks, four model backbones, and a college-level analytical writing study shows that MetaCrit significantly improves content truthfulness and logical soundness while eliminating toxic outputs. Its modular design allows individual agents to be integrated into existing frameworks as drop-in components without architectural modifications.

Key facts. Cited by 1 paper 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 11 of its 37 references that are in the map. First posted on arXiv 20 Jul 2025.

Builds on

11 of 37 references.

References · 11

  1. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  2. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,330
  3. 2022Large Language Models are Zero-Shot Reasonerscited by 4,374
  4. 2023Improving Factuality and Reasoning in Language Models through Multiagent Debatecited by 1,784
  5. 2023Chain-of-Verification Reduces Hallucination in Large Language Modelscited by 324
  6. 2023Universal Self-Consistency for Large Language Model Generationcited by 203
  7. 2023Take a Step Back: Evoking Reasoning via Abstraction in Large Language Modelscited by 159
  8. 2023ExpertPrompting: Instructing Large Language Models to be Distinguished Expertscited by 142
  9. 2025Benchmarking Adversarial Robustness to Bias Elicitation in Large Language Models: Scalable Automated Assessment with LLM-as-a-Judgecited by 23
  10. 2025ReflectEvo: Improving Meta Introspection of Small LLMs by Learning Self-Reflectioncited by 12
  11. 2025Can LLMs Reconcile Knowledge Conflicts in Counterfactual Reasoningcited by 7

Built on by

Cited by year

Citing papers per year: 2025: 1.

1 paper.

  1. 2025Perspectra: Choosing Your Experts Enhances Critical Thinking in Multi-Agent Research Ideationcited by 4

Similar, not cited

Close in wording, no citation either way

  1. 2023Encouraging Divergent Thinking in Large Language Models through Multi-Agent Debate
  2. 2026Think2: Grounded Metacognitive Reasoning in Large Language Models
  3. 2026Meta-Moderator: Empowering Multi-Agent Debate with Meta-Cognition
  4. 2026Conversation for Non-verifiable Learning: Self-Evolving LLMs through Meta-Evaluation
  5. 2026The Reasoning Trap: An Information-Theoretic Bound on Closed-System Multi-Step LLM Reasoning
  6. 2026Metacognitive Behavioral Tuning of Large Language Models for Multi-Hop Question Answering
  7. 2025Stepwise Think-Critique: Interleaved Reasoning and Self-Critique in a Single LLM
  8. 2024Diversity of Thought Elicits Stronger Reasoning Capabilities in Multi-Agent Debate Frameworks