RankT5: Fine-Tuning T5 for Text Ranking with Ranking Losses

• URL: http://arxiv.org/abs/2210.10634v1
• Date: Wed, 12 Oct 2022 20:51:49 GMT
• Title: RankT5: Fine-Tuning T5 for Text Ranking with Ranking Losses
• Authors: Honglei Zhuang, Zhen Qin, Rolf Jagerman, Kai Hui, Ji Ma, Jing Lu, Jianmo Ni, Xuanhui Wang and Michael Bendersky
• Abstract summary: We propose two T5-based ranking model structures, an encoder-decoder and an encoder-only one, so that they can directly output ranking scores for each query-document pair. Our experiments show that the proposed models with ranking losses can achieve substantial ranking performance gains on different public text ranking data sets.
• Score: 39.67403439576671
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recently, substantial progress has been made in text ranking based on pretrained language models such as BERT. However, there are limited studies on how to leverage more powerful sequence-to-sequence models such as T5. Existing attempts usually formulate text ranking as classification and rely on postprocessing to obtain a ranked list. In this paper, we propose RankT5 and study two T5-based ranking model structures, an encoder-decoder and an encoder-only one, so that they not only can directly output ranking scores for each query-document pair, but also can be fine-tuned with "pairwise" or "listwise" ranking losses to optimize ranking performances. Our experiments show that the proposed models with ranking losses can achieve substantial ranking performance gains on different public text ranking data sets. Moreover, when fine-tuned with listwise ranking losses, the ranking model appears to have better zero-shot ranking performance on out-of-domain data sets compared to the model fine-tuned with classification losses.

Related papers

• Efficient course recommendations with T5-based ranking and summarization [2.6968321526169503]
  We develop a two-step retrieval pipeline with RankT5 finetuned on MSMARCO as re-ranker. We evaluate our rankers on two newly labelled datasets, with an A/B test, and with a user questionnaire. We conclude that T5-based re-ranking and summarization for online course recommendation can obtain much better effectiveness than single-step lexical retrieval.
  arXiv  Detail & Related papers  (2024-06-27T09:07:32Z)
• Found in the Middle: Permutation Self-Consistency Improves Listwise Ranking in Large Language Models [63.714662435555674]
  Large language models (LLMs) exhibit positional bias in how they use context. We propose permutation self-consistency, a form of self-consistency over ranking list outputs of black-box LLMs. Our approach improves scores from conventional inference by up to 7-18% for GPT-3.5 and 8-16% for LLaMA v2 (70B)
  arXiv  Detail & Related papers  (2023-10-11T17:59:02Z)
• Replace Scoring with Arrangement: A Contextual Set-to-Arrangement Framework for Learning-to-Rank [40.81502990315285]
  Learning-to-rank is a core technique in the top-N recommendation task, where an ideal ranker would be a mapping from an item set to an arrangement. Most existing solutions fall in the paradigm of probabilistic ranking principle (PRP), i.e., first score each item in the candidate set and then perform a sort operation to generate the top ranking list. We propose Set-To-Arrangement Ranking (STARank), a new framework directly generates the permutations of the candidate items without the need for individually scoring and sort operations.
  arXiv  Detail & Related papers  (2023-08-05T12:22:26Z)
• Learning to Rank when Grades Matter [11.981942948477236]
  Graded labels are ubiquitous in real-world learning-to-rank applications. Traditional learning-to-rank techniques ignore predicting actual grades. We propose a multiobjective formulation to jointly optimize both ranking and grade predictions.
  arXiv  Detail & Related papers  (2023-06-14T17:30:02Z)
• Zero-Shot Listwise Document Reranking with a Large Language Model [58.64141622176841]
  We propose Listwise Reranker with a Large Language Model (LRL), which achieves strong reranking effectiveness without using any task-specific training data. Experiments on three TREC web search datasets demonstrate that LRL not only outperforms zero-shot pointwise methods when reranking first-stage retrieval results, but can also act as a final-stage reranker.
  arXiv  Detail & Related papers  (2023-05-03T14:45:34Z)
• T5Score: Discriminative Fine-tuning of Generative Evaluation Metrics [94.69907794006826]
  We present a framework that combines the best of both worlds, using both supervised and unsupervised signals from whatever data we have available. We operationalize this idea by training T5Score, a metric that uses these training signals with mT5 as the backbone. T5Score achieves the best performance on all datasets against existing top-scoring metrics at the segment level.
  arXiv  Detail & Related papers  (2022-12-12T06:29:04Z)
• Differentiable Top-k Classification Learning [29.75063301688965]
  We optimize the model for multiple k simultaneously instead of using a single k. We find that relaxing k does not only produce better top-5 accuracies, but also leads to top-1 accuracy improvements.
  arXiv  Detail & Related papers  (2022-06-15T04:13:59Z)
• PiRank: Learning To Rank via Differentiable Sorting [85.28916333414145]
  We propose PiRank, a new class of differentiable surrogates for ranking. We show that PiRank exactly recovers the desired metrics in the limit of zero temperature.
  arXiv  Detail & Related papers  (2020-12-12T05:07:36Z)
• Rank over Class: The Untapped Potential of Ranking in Natural Language Processing [8.637110868126546]
  We argue that many tasks which are currently addressed using classification are in fact being shoehorned into a classification mould. We propose a novel end-to-end ranking approach consisting of a Transformer network responsible for producing representations for a pair of text sequences. In an experiment on a heavily-skewed sentiment analysis dataset, converting ranking results to classification labels yields an approximately 22% improvement over state-of-the-art text classification.
  arXiv  Detail & Related papers  (2020-09-10T22:18:57Z)
• Document Ranking with a Pretrained Sequence-to-Sequence Model [56.44269917346376]
  We show how a sequence-to-sequence model can be trained to generate relevance labels as "target words" Our approach significantly outperforms an encoder-only model in a data-poor regime.
  arXiv  Detail & Related papers  (2020-03-14T22:29:50Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.