![[2602.20735] RMIT-ADM+S at the MMU-RAG NeurIPS 2025 Competition](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2602.20735] RMIT-ADM+S at the MMU-RAG NeurIPS 2025 Competition
Summary
The paper presents RMIT-ADM+S, an award-winning system for the Text-to-Text track at the NeurIPS 2025 Competition, featuring a novel retrieval-augmented generation architecture called Routing-to-RAG (R2RAG).
Why It Matters
This research is significant as it showcases advancements in retrieval-augmented generation systems, particularly in optimizing performance on consumer-grade hardware. The RMIT-ADM+S system's success highlights the potential for efficient AI solutions in complex tasks, making it relevant for both academic and practical applications in AI and information retrieval.
Key Takeaways
- RMIT-ADM+S won the Best Dynamic Evaluation award at NeurIPS 2025.
- The Routing-to-RAG (R2RAG) architecture adapts retrieval strategies based on query complexity.
- The system operates efficiently on a single consumer-grade GPU.
- It builds upon the successful G-RAG system, enhancing its capabilities.
- The design focuses on effective resource use while maintaining high performance.
Computer Science > Information Retrieval arXiv:2602.20735 (cs) [Submitted on 24 Feb 2026] Title:RMIT-ADM+S at the MMU-RAG NeurIPS 2025 Competition Authors:Kun Ran, Marwah Alaofi, Danula Hettiachchi, Chenglong Ma, Khoi Nguyen Dinh Anh, Khoi Vo Nguyen, Sachin Pathiyan Cherumanal, Lida Rashidi, Falk Scholer, Damiano Spina, Shuoqi Sun, Oleg Zendel View a PDF of the paper titled RMIT-ADM+S at the MMU-RAG NeurIPS 2025 Competition, by Kun Ran and 11 other authors View PDF HTML (experimental) Abstract:This paper presents the award-winning RMIT-ADM+S system for the Text-to-Text track of the NeurIPS~2025 MMU-RAG Competition. We introduce Routing-to-RAG (R2RAG), a research-focused retrieval-augmented generation (RAG) architecture composed of lightweight components that dynamically adapt the retrieval strategy based on inferred query complexity and evidence sufficiency. The system uses smaller LLMs, enabling operation on a single consumer-grade GPU while supporting complex research tasks. It builds on the G-RAG system, winner of the ACM~SIGIR~2025 LiveRAG Challenge, and extends it with modules informed by qualitative review of outputs. R2RAG won the Best Dynamic Evaluation award in the Open Source category, demonstrating high effectiveness with careful design and efficient use of resources. Comments: Subjects: Information Retrieval (cs.IR); Artificial Intelligence (cs.AI); Computation and Language (cs.CL) Cite as: arXiv:2602.20735 [cs.IR] (or arXiv:2602.20735v1 [cs.IR] for this vers...