MedVerse: Efficient and Reliable Medical Reasoning via DAG-Structured Parallel Execution

ACL ARR 2026 January Submission5917 Authors

Keywords: Medical LLM, LLM Reasoning, Parallel Generation

Abstract:

Large language models (LLMs) have demonstrated strong performance and rapid progress in a wide range of medical reasoning tasks.
However, their sequential autoregressive decoding forces inherently parallel clinical reasoning, such as differential diagnosis, into a single linear reasoning path, limiting both efficiency and reliability for complex medical problems.
To address this, we propose MedVerse, a reasoning framework for complex medical inference that reformulates medical reasoning as a parallelizable directed acyclic graph (DAG) process based on Petri Net theory.
The framework adopts a full-stack design across data, model architecture, and system execution.
For data creation, we introduce the MedVerse Curator, an automated pipeline that synthesizes knowledge-grounded medical reasoning path and transforms them into Petri Net–structured representations.
At the architectural level, we propose a topology-aware attention mechanism with adaptive position indices that supports parallel reasoning while preserving logical consistency.
Systematically, we develop a customized inference engine that supports memory-efficient parallel execution.
Empirical evaluations show that MedVerse improves strong general-purpose LLMs by up to 8.9\%. Compared to specialized medical LLMs, MedVerse achieves comparable performance while delivering a 1.3$\times$ reduction in inference latency and a 1.7$\times$ increase in generation throughput, enabled by its parallel decoding capability.

Paper Type: Long

Research Area: Clinical and Biomedical Applications

Research Area Keywords: Clinical and Biomedical Applications, Generation, Language Modeling, NLP Applications, Question Answering

Contribution Types: NLP engineering experiment, Approaches low compute settings-efficiency

Languages Studied: English

Submission Number: 5917