HLTCOE at LiveRAG: GPT-Researcher using ColBERT retrieval

2025/06/27 by Kevin Duh, Eugene Yang, Duh, Kevin +7 · 1 citation
Computer Science · Medicine · #Artificial Intelligence in Healthcare and Education #FOS: Computer and information sciences #Information Retrieval (cs.IR) #Mathematics, Computing, and Information Processing #Scientific Research and Technology

paper · pdf · doi:10.48550/arxiv.2506.22356

openalex publication_date 2025/06/27 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

The HLTCOE LiveRAG submission utilized the GPT-researcher framework for researching the context of the question, filtering the returned results, and generating the final answer. The retrieval system was a ColBERT bi-encoder architecture, which represents a passage with many dense tokens. Retrieval used a local, compressed index of the FineWeb10-BT collection created with PLAID-X, using a model fine-tuned for multilingual retrieval. Query generation from context was done with Qwen2.5-7B-Instruct, while filtering was accomplished with m2-bert-80M-8k-retrieval. Up to nine passages were used as context to generate an answer using Falcon3-10B. This system placed 5th in the LiveRAG automatic evaluation for correctness with a score of 1.07.

Citations

• Transfer Learning Approaches for Building Cross-Language Dense Retrieval Models
• The Sockeye 2 Neural Machine Translation Toolkit at AMTA 2020
• Unsupervised Cross-lingual Representation Learning at Scale

Cited by

• SIGIR 2025 -- LiveRAG Challenge Report

Related

• ModernBERT + ColBERT: Enhancing biomedical RAG through an advanced re-ranking retriever
• SIGIR 2025 -- LiveRAG Challenge Report
• TopClustRAG at SIGIR 2025 LiveRAG Challenge
• ColBERT: Efficient and Effective Passage Search via Contextualized Late Interaction over BERT
• LiveRAG: A diverse Q&A dataset with varying difficulty level for RAG evaluation