TREC 2025 Proceedings

Submission Details

Organization

dgthesis

Track

Tip-of-the-Tongue Search

Task

Retrieval Task

Date

2025-09-10

Run Description

Please describe in details how this run was generated

Building on the pyterrier bm25 baseline run of the test set provided by the organizing team, I first extracted all documents that appear in that run into a smaller corpus dataset to rerank with. Using the reranker module and the 'mixedbread-ai/mxbai-rerank-large-v1' model I then reranked the 1000 documents for each query, directly outputting each query's reranking results to the runfile.

Specify datasets used in this run.

["This year's TREC TOT training data", 'Other']

(if you checked "other", describe here)

The model used (mixedbread-ai/mxbai-rerank-large-v1) was presumably trained using other data, but I don't know exactly what.

Are you 100% confident that no data from https://github.com/microsoft/Tip-of-the-Tongue-Known-Item-Retrieval-Dataset-for-Movie-Identification or iRememberThisMovie.com (besides the training data provided as part of this year's track) was used for producing this run (including any data used for pretraining models that you are building on top of)?

no

Did you use any of the official baseline runs in any way to produce this run?

yes

If you did use any of the official baseline runs in any way to produce this run, please describe how below in sufficient detail (e.g., as reranking candidates or in ensemble with other approaches).

I used the runfile produced by the BM25 PyTerrier baseline as a starting point for reranking. I chose this baseline because it had the best R@1000 for the dev3 dataset, which is the only multi-domain set available and thus most relevant compared to the test set.

Evaluation Files

• dgMxbaiL01.trec_eval (trec_eval)