Overview of the TREC 2023 deep learning track

TL;DR

The paper reviews the TREC 2023 deep learning track, highlighting the use of large language models for ranking tasks, the creation of synthetic queries, and the performance of prompt-based methods surpassing previous approaches.

Contribution

It introduces the use of LLM prompting in ranking tasks, compares synthetic and human queries, and demonstrates the effectiveness of prompt-based methods in the TREC 2023 track.

Findings

LLM prompting outperformed nnlm approaches.

Synthetic queries yielded similar system rankings as human queries.

No clear bias observed between GPT-4 and T5 evaluations.

Abstract

This is the fifth year of the TREC Deep Learning track. As in previous years, we leverage the MS MARCO datasets that made hundreds of thousands of human-annotated training labels available for both passage and document ranking tasks. We mostly repeated last year's design, to get another matching test set, based on the larger, cleaner, less-biased v2 passage and document set, with passage ranking as primary and document ranking as a secondary task (using labels inferred from passage). As we did last year, we sample from MS MARCO queries that were completely held out, unused in corpus construction, unlike the test queries in the first three years. This approach yields a more difficult test with more headroom for improvement. Alongside the usual MS MARCO (human) queries from MS MARCO, this year we generated synthetic queries using a fine-tuned T5 model and using a GPT-4 prompt. The new headline result this year is that runs using Large Language Model (LLM) prompting in some way outperformed runs that use the "nnlm" approach, which was the best approach in the previous four years. Since this is the last year of the track, future iterations of prompt-based ranking can happen in other tracks. Human relevance assessments were applied to all query types, not just human MS MARCO queries. Evaluation using synthetic queries gave similar results to human queries, with system ordering agreement of $\tau=0.8487$. However, human effort was needed to select a subset of the synthetic queries that were usable. We did not see clear evidence of bias, where runs using GPT-4 were favored when evaluated using synthetic GPT-4 queries, or where runs using T5 were favored when evaluated on synthetic T5 queries.