# HLTCOE at LiveRAG: GPT-Researcher using ColBERT retrieval

arXiv: https://arxiv.org/abs/2506.22356 · HTML: https://mlsanity.com/paper/2506.22356/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 27 Jun 2025.

## Built on by (most cited first, 6 of 6)

- [Incorporating Q&A Nuggets into Retrieval-Augmented Generation](https://mlsanity.com/paper/2601.13222/) (2026)
- [LANCER: LLM Reranking for Nugget Coverage](https://mlsanity.com/paper/2601.22008/) (2026)
- [RoutIR: Fast Serving of Retrieval Pipelines for Retrieval-Augmented Generation](https://mlsanity.com/paper/2601.10644/) (2026)
- [Insider Knowledge: How Much Can RAG Systems Gain from Evaluation Secrets?](https://mlsanity.com/paper/2601.13227/) (2026)
- [Beyond Relevance: On the Relationship Between Retrieval and RAG Information Coverage](https://mlsanity.com/paper/2603.08819/) (2026)
- [SIGIR 2025 -- LiveRAG Challenge Report](https://mlsanity.com/paper/2507.04942/) (2025)

## Abstract (arXiv)

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.