MedCoG: Maximizing LLM Inference Density in Medical Reasoning via Meta-Cognitive Regulation
Yu Zhao, Hao Guan, Yongcheng Jing, Ying Zhang, Dacheng Tao
ICML 2026 regular
Abstract (source: OpenReview · © authors)
Large Language Models (LLMs) have shown strong potential in complex medical reasoning yet face diminishing gains under inference scaling laws. While existing studies augment LLMs with various knowledge types, it remains unclear how effectively the additional costs translate into accuracy. In this paper, we explore how meta-cognition of LLMs, i.e., their self-assessment of their own cognitive states, can regulate the reasoning process. Specifically, we propose MedCoG, a Medical Meta-Cognition Agent with Knowledge Graph, where the meta-cognitive assessments of task complexity, familiarity, and knowledge density dynamically regulate utilization of procedural, episodic, and factual knowledge. The LLM-centric on-demand reasoning aims to mitigate the diminishing returns under scaling law by (1) reducing costs via avoiding indiscriminate scaling, (2) improving accuracy via filtering out distractive knowledge. To validate this, we empirically characterize the scaling curve and introduce inference density to quantify inference efficiency. Experiments demonstrate the effectiveness and efficiency of MedCoG on five hard sets of medical benchmarks, yielding 6.2x inference density. Furthermore, the Oracle study highlights the significant potential of meta-cognitive regulation.
Keywords
Metadata from BioTender-max/icml2026-ai-bio (CC0-1.0). Phở does not host any PDF; links point back to the source.
Related
From Conflict to Consensus: Boosting Medical Reasoning via Multi-Round Agentic RAG
Wenhao Wu, Zhentao Tang, Yafu Li, Shixiong Kai +4
Large Language Models (LLMs) exhibit high reasoning capacity in medical question-answering, but their tendency to produce hallucinations and outdated knowledge poses critical…
Expert-guided Clinical Text Augmentation via Query-Based Model Collaboration
Dongkyu Cho, Miao Zhang, Gregory D Lyng, Rumi Chunara
Data augmentation is a widely used strategy to improve model robustness and generalization by enriching training datasets with synthetic examples. While large language models…
Influence-Guided Symbolic Regression: Scientific Discovery via LLM-Driven Equation Search with Granular Feedback
Evgeny Saveliev, Samuel Holt, Nabeel Seedat, David L. Bentley +2
Large Language Models (LLMs) offer a promising avenue for scientific discovery, yet their application to symbolic regression is often constrained by inefficient search strategies…