DoTA-RAG: Dynamic of Thought Aggregation RAG

TL;DR

DoTA-RAG is a retrieval-augmented generation system designed for fast, accurate access to large-scale web knowledge, improving answer correctness and efficiency through dynamic routing and optimized embedding models.

Contribution

The paper introduces DoTA-RAG, a novel three-stage retrieval pipeline with dynamic routing and enhanced embedding evaluation, enabling high-throughput, accurate web knowledge retrieval.

Findings

Answer correctness score improved from 0.752 to 1.478.

Achieved a 0.929 correctness score on the Live Challenge Day.

Maintains low latency while handling large, diverse datasets.

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.