An In-depth Analysis Of Passage-level Label Transfer For Contextual Document Ranking

Abstract

Pre-trained contextual language models such as BERT, GPT, and XLnet work quite well for document retrieval tasks. Such models are fine-tuned based on the query-document/query-passage level relevance labels to capture the ranking signals. However, the documents are longer than the passages and such document ranking models suffer from the token limitation (512) of BERT. Researchers proposed ranking strategies that either truncate the documents beyond the token limit or chunk the documents into units that can fit into the BERT. In the later case, the relevance labels are either directly transferred from the original query-document pair or learned through some external model. In this paper, we conduct a detailed study of the design decisions about splitting and label transfer on retrieval effectiveness and efficiency. We find that direct transfer of relevance labels from documents to passages introduces label noise that strongly affects retrieval effectiveness for large training datasets.

Stats

• citations1
• S2 citations—
• github stars0
• HF likes0
• heat score2.26
• arxiv keyrudra2021an

Related papers

• A Passage-based Approach To Learning To Rank Documents (2019)8.60
• Colbert: Efficient And Effective Passage Search Via Contextualized Late Interaction Over BERT (2020)0.00
• How Different Are Pre-trained Transformers For Text Ranking? (2022)7.81
• Query-as-context Pre-training For Dense Passage Retrieval (2022)7.68
• Context Is Gold To Find The Gold Passage: Evaluating And Training Contextual Document Embeddings (2025)5.62
• CODER: An Efficient Framework For Improving Retrieval Through Contextual Document Embedding Reranking (2021)7.16
• Pre-training Tasks For Embedding-based Large-scale Retrieval (2020)0.00
• On The Interpolation Of Contextualized Term-based Ranking With BM25 For Query-by-example Retrieval (2022)7.50