# Content-Based Weak Supervision for Ad-Hoc Re-Ranking

arXiv: https://arxiv.org/abs/1707.00189 · HTML: https://mlsanity.com/paper/1707.00189/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 1 Jul 2017.

## Built on by (most cited first, 7 of 7)

- [Simplified Data Wrangling with ir_datasets](https://mlsanity.com/paper/2103.02280/) (2021)
- [Low-Resource Dense Retrieval for Open-Domain Question Answering: A Comprehensive Survey](https://mlsanity.com/paper/2208.03197/) (2022)
- [EmbedDistill: A Geometric Knowledge Distillation for Information Retrieval](https://mlsanity.com/paper/2301.12005/) (2023)
- [Learnable Pillar-based Re-ranking for Image-Text Retrieval](https://mlsanity.com/paper/2304.12570/) (2023)
- [RANKVIDEO: Reasoning Reranking for Text-to-Video Retrieval](https://mlsanity.com/paper/2602.02444/) (2026)
- [Improving Neural Ranking Models with Traditional IR Methods](https://mlsanity.com/paper/2308.15027/) (2023)
- [Domain Adaptation of Multilingual Semantic Search -- Literature Review](https://mlsanity.com/paper/2402.02932/) (2024)

## Abstract (arXiv)

One challenge with neural ranking is the need for a large amount of manually-labeled relevance judgments for training. In contrast with prior work, we examine the use of weak supervision sources for training that yield pseudo query-document pairs that already exhibit relevance (e.g., newswire headline-content pairs and encyclopedic heading-paragraph pairs). We also propose filtering techniques to eliminate training samples that are too far out of domain using two techniques: a heuristic-based approach and novel supervised filter that re-purposes a neural ranker. Using several leading neural ranking architectures and multiple weak supervision datasets, we show that these sources of training pairs are effective on their own (outperforming prior weak supervision techniques), and that filtering can further improve performance.