Single-Turn LLM Reformulation Powered Multi-Stage Hybrid Re-Ranking for Tip-of-the-Tongue Known-Item Retrieval

TL;DR

This paper introduces a multi-stage hybrid re-ranking approach for Tip-of-the-Tongue known-item retrieval, leveraging a single LLM call for query reformulation to significantly improve retrieval performance without model fine-tuning.

Contribution

It presents a novel prompt-based query reformulation method using an 8B-parameter LLM, enhancing retrieval effectiveness in ToT scenarios across multiple reranking stages.

Findings

20.61% increase in recall with query reformulation

33.88% improvement in nDCG@10 after reranking

29.92% boost in MRR through multi-stage reranking

Abstract

Retrieving known items from vague descriptions, Tip-of-the-Tongue (ToT) retrieval, remains a significant challenge. We propose using a single call to a generic 8B-parameter LLM for query reformulation, bridging the gap between ill-formed ToT queries and specific information needs. This method is particularly effective where standard Pseudo-Relevance Feedback fails due to poor initial recall. Crucially, our LLM is not fine-tuned for ToT or specific domains, demonstrating that gains stem from our prompting strategy rather than model specialization. Rewritten queries feed a multi-stage pipeline: sparse retrieval (BM25), dense/late-interaction reranking (Contriever, E5-large-v2, ColBERTv2), monoT5 cross-encoding, and list-wise reranking (Qwen 2.5 72B). Experiments on 2025 TREC-ToT datasets show that while raw queries yield poor performance, our lightweight pre-retrieval transformation improves Recall by 20.61%. Subsequent reranking improves nDCG@10 by 33.88%, MRR by 29.92%, and MAP@10 by 29.98%, offering a cost-effective intervention that unlocks the potential of downstream rankers. Code and data: https://github.com/debayan1405/TREC-TOT-2025