# Semi-Siamese Bi-encoder Neural Ranking Model Using Lightweight Fine-Tuning

arXiv: https://arxiv.org/abs/2110.14943 · HTML: https://mlsanity.com/paper/2110.14943/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 26 references that are in the map.
- First posted on arXiv 28 Oct 2021.

## Builds on (most cited first, 19 of 19)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Parameter-Efficient Transfer Learning for NLP](https://mlsanity.com/paper/1902.00751/) (2019)
- [Prefix-Tuning: Optimizing Continuous Prompts for Generation](https://mlsanity.com/paper/2101.00190/) (2021)
- [The Power of Scale for Parameter-Efficient Prompt Tuning](https://mlsanity.com/paper/2104.08691/) (2021)
- [Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processing](https://mlsanity.com/paper/2107.13586/) (2021)
- [MS MARCO: A Human Generated MAchine Reading COmprehension Dataset](https://mlsanity.com/paper/1611.09268/) (2016)
- [BitFit: Simple Parameter-efficient Fine-tuning for Transformer-based Masked Language-models](https://mlsanity.com/paper/2106.10199/) (2021)
- [ColBERT: Efficient and Effective Passage Search via Contextualized Late Interaction over BERT](https://mlsanity.com/paper/2004.12832/) (2020)
- [Passage Re-ranking with BERT](https://mlsanity.com/paper/1901.04085/) (2019)
- [Revisiting Few-sample BERT Fine-tuning](https://mlsanity.com/paper/2006.05987/) (2020)
- [On the Effectiveness of Adapter-based Tuning for Pretrained Language Model Adaptation](https://mlsanity.com/paper/2106.03164/) (2021)
- [What Would Elsa Do? Freezing Layers During Transformer Fine-Tuning](https://mlsanity.com/paper/1911.03090/) (2019)
- [CEDR: Contextualized Embeddings for Document Ranking](https://mlsanity.com/paper/1904.07094/) (2019)
- [AutoFreeze: Automatically Freezing Model Blocks to Accelerate Fine-tuning](https://mlsanity.com/paper/2102.01386/) (2021)
- [How fine can fine-tuning be? Learning efficient language models](https://mlsanity.com/paper/2004.14129/) (2020)
- [TwinBERT: Distilling Knowledge to Twin-Structured BERT Models for Efficient Retrieval](https://mlsanity.com/paper/2002.06275/) (2020)
- [Semi-Siamese Training for Shallow Face Learning](https://mlsanity.com/paper/2007.08398/) (2020)

## Built on by (most cited first, 6 of 6)

- [Isotropic Representation Can Improve Dense Retrieval](https://mlsanity.com/paper/2209.00218/) (2022)
- [Disentangled Modeling of Domain and Relevance for Adaptable Dense Retrieval](https://mlsanity.com/paper/2208.05753/) (2022)
- [PEFA: Parameter-Free Adapters for Large-scale Embedding-based Retrieval Models](https://mlsanity.com/paper/2312.02429/) (2023)
- [Query-driven Relevant Paragraph Extraction from Legal Judgments](https://mlsanity.com/paper/2404.00595/) (2024)
- [Parameter-Efficient Sparse Retrievers and Rerankers using Adapters](https://mlsanity.com/paper/2303.13220/) (2023)
- [CREST: Improving Interpretability and Effectiveness of Troubleshooting at Ericsson through Criterion-Specific Trouble Report Retrieval](https://mlsanity.com/paper/2511.17417/) (2025)

## Abstract (arXiv)

A BERT-based Neural Ranking Model (NRM) can be either a crossencoder or a bi-encoder. Between the two, bi-encoder is highly efficient because all the documents can be pre-processed before the actual query time. In this work, we show two approaches for improving the performance of BERT-based bi-encoders. The first approach is to replace the full fine-tuning step with a lightweight fine-tuning. We examine lightweight fine-tuning methods that are adapter-based, prompt-based, and hybrid of the two. The second approach is to develop semi-Siamese models where queries and documents are handled with a limited amount of difference. The limited difference is realized by learning two lightweight fine-tuning modules, where the main language model of BERT is kept common for both query and document. We provide extensive experiment results for monoBERT, TwinBERT, and ColBERT where three performance metrics are evaluated over Robust04, ClueWeb09b, and MS-MARCO datasets. The results confirm that both lightweight fine-tuning and semi-Siamese are considerably helpful for improving BERT-based bi-encoders. In fact, lightweight fine-tuning is helpful for crossencoder, too