EUVIMEDEUVIMED entdecken
Crossref

MIRAGE: Scaling Test-Time Inference with Parallel Graph-Retrieval-Augmented Reasoning Chains

Kaiwen Wei, Rui Shan, Dongsheng Zou, Jianzhong Yang, Bi Zhao, Junnan Zhu, Jiang Zhong

Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (40) · 33818-33826

DOI: 10.1609/aaai.v40i40.40673 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Kaiwen Wei und 6 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Large reasoning models (LRMs) have shown significant progress in test-time scaling through chain-of-thought prompting. Current approaches like search-o1 integrate retrieval augmented generation (RAG) into multi-step reasoning processes but rely on a single, linear reasoning path while incorporating unstructured textual information in a flat, context-agnostic manner. As a result, these approaches can lead to error accumulation throughout the reasoning chain, which significantly limits its effectiveness in medical question-answering (QA) tasks where both accuracy and traceability are critical requirements. To address these challenges, we propose MIRAGE (Multi-path Inference with Retrieval-Augmented Graph Exploration), a novel test-time scalable reasoning framework that performs dynamic multi-path inference over structured medical knowledge graphs. Specifically, MIRAGE 1) decomposes complex queries into entity-grounded sub-questions, 2) executes parallel inference paths, 3) retrieves evidence adaptively via neighbor expansion and multi-hop traversal, and 4) integrates answers using cross-path verification to resolve contradictions. Experiments on three medical QA benchmarks (GenMedGPT-5k, CMCQA, and ExplainCPE) show that MIRAGE consistently outperforms GPT-4o, Tree-of-Thought variants, and other retrieval-augmented baselines in both automatic and human evaluations. Additionally, MIRAGE improves interpretability by generating explicit reasoning chains that trace each factual claim to concrete paths within the knowledge graph, making it especially suitable for complex medical reasoning scenarios.

Zitieren

APA

Wei, K., Shan, R., Zou, D., Yang, J., Zhao, B., Zhu, J., & Zhong, J. (2026). MIRAGE: Scaling Test-Time Inference with Parallel Graph-Retrieval-Augmented Reasoning Chains. Proceedings of the AAAI Conference on Artificial Intelligence, 40(40), 33818-33826. https://doi.org/10.1609/aaai.v40i40.40673

Vancouver

Wei K, Shan R, Zou D, Yang J, Zhao B, Zhu J, et al. MIRAGE: Scaling Test-Time Inference with Parallel Graph-Retrieval-Augmented Reasoning Chains. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(40):33818-33826. doi:10.1609/aaai.v40i40.40673

BibTeX
@article{Wei2026,
  author = {Wei, Kaiwen and Shan, Rui and Zou, Dongsheng and Yang, Jianzhong and Zhao, Bi and Zhu, Junnan and Zhong, Jiang},
  title = {MIRAGE: Scaling Test-Time Inference with Parallel Graph-Retrieval-Augmented Reasoning Chains},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2026},
  volume = {40},
  number = {40},
  pages = {33818--33826},
  doi = {10.1609/aaai.v40i40.40673}
}
Kostenlos starten →