DoTA-RAG: Dynamic of Thought Aggregation RAG

2025/06/14 by Saksorn Ruangtanusak, Natthapath Rungseesiripak, Ruangtanusak, Saksorn +7 · 1 citation
Neuroscience · #Artificial Intelligence (cs.AI) #Computation and Language (cs.CL) #FOS: Computer and information sciences #Functional Brain Connectivity Studies #Information Retrieval (cs.IR)

paper · pdf · doi:10.48550/arxiv.2506.12571

openalex publication_date 2025/06/14 · openalex created_date 2025/10/13 · openalex updated_date 2026/07/28

Abstract

In this paper, we introduce DoTA-RAG (Dynamic-of-Thought Aggregation RAG), a retrieval-augmented generation system optimized for high-throughput, large-scale web knowledge indexes. Traditional RAG pipelines often suffer from high latency and limited accuracy over massive, diverse datasets. DoTA-RAG addresses these challenges with a three-stage pipeline: query rewriting, dynamic routing to specialized sub-indexes, and multi-stage retrieval and ranking. We further enhance retrieval by evaluating and selecting a superior embedding model, re-embedding the large FineWeb-10BT corpus. Moreover, we create a diverse Q&A dataset of 500 questions generated via the DataMorgana setup across a broad range of WebOrganizer topics and formats. DoTA-RAG improves the answer correctness score from 0.752 (baseline, using LiveRAG pre-built vector store) to 1.478 while maintaining low latency, and it achieves a 0.929 correctness score on the Live Challenge Day. These results highlight DoTA-RAG's potential for practical deployment in domains requiring fast, reliable access to large and evolving knowledge sources.

Citations

• Chain of Draft: Thinking Faster by Writing Less
• Organize the Web: Constructing Domains Enhances Pre-Training Data Curation
• Generating Diverse Q&A Benchmarks for RAG Evaluation with DataMorgana
• Arctic-Embed 2.0: Multilingual Retrieval Without Compromise
• The FineWeb Datasets: Decanting the Web for the Finest Text Data at Scale
• Retrieval-Augmented Generation for Large Language Models: A Survey
• From Classification to Generation: Insights into Crosslingual Retrieval Augmented ICL
• Precise Zero-Shot Dense Retrieval without Relevance Labels
• Text Embeddings by Weakly-Supervised Contrastive Pre-training
• Self-Consistency Improves Chain of Thought Reasoning in Language Models
• ColBERT: Efficient and Effective Passage Search via Contextualized Late Interaction over BERT

Cited by

• SIGIR 2025 -- LiveRAG Challenge Report

Related

• RAG in the Wild: On the (In)effectiveness of LLMs with Mixture-of-Knowledge Retrieval Augmentation
• CRAG -- Comprehensive RAG Benchmark
• Zero-RAG: Towards Retrieval-Augmented Generation with Zero Redundant Knowledge
• TopClustRAG at SIGIR 2025 LiveRAG Challenge
• Context-Guided Dynamic Retrieval for Improving Generation Quality in RAG Models