Generative Relevance Feedback And Convergence Of Adaptive Re-ranking: University Of Glasgow Terrier Team At TREC DL 2023

Abstract

This paper describes our participation in the TREC 2023 Deep Learning Track. We submitted runs that apply generative relevance feedback from a large language model in both a zero-shot and pseudo-relevance feedback setting over two sparse retrieval approaches, namely BM25 and SPLADE. We couple this first stage with adaptive re-ranking over a BM25 corpus graph scored using a monoELECTRA cross-encoder. We investigate the efficacy of these generative approaches for different query types in first-stage retrieval. In re-ranking, we investigate operating points of adaptive re-ranking with different first stages to find the point in graph traversal where the first stage no longer has an effect on the performance of the overall retrieval pipeline. We find some performance gains from the application of generative query reformulation. However, our strongest run in terms of P@10 and nDCG@10 applied both adaptive re-ranking and generative pseudo-relevance feedback, namely uogtr_b_grf_e_gb.

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyparry2024generative

Related papers

• Generative Retrieval Meets Multi-graded Relevance (2024)2.26
• A Deep Learning Approach For Selective Relevance Feedback (2024)6.34
• DS@GT At TREC TOT 2025: Bridging Vague Recollection With Fusion Retrieval And Learned Reranking (2026)0.00
• PASH At TREC 2021 Deep Learning Track: Generative Enhanced Model For Multi-stage Ranking (2022)0.00
• Naver Labs Europe (SPLADE) @ TREC Deep Learning 2022 (2023)0.00
• Improving Query Representations For Dense Retrieval With Pseudo Relevance Feedback: A Reproducibility Study (2021)7.16
• Refit: Relevance Feedback From A Reranker During Inference (2023)0.00
• Learning To Rank In Generative Retrieval (2023)11.91