# Cascaded Fast and Slow Models for Efficient Semantic Code Search

arXiv: https://arxiv.org/abs/2110.07811 · HTML: https://mlsanity.com/paper/2110.07811/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 24 references that are in the map.
- First posted on arXiv 15 Oct 2021.

## Builds on (most cited first, 20 of 20)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Program Synthesis with Large Language Models](https://mlsanity.com/paper/2108.07732/) (2021)
- [Align before Fuse: Vision and Language Representation Learning with Momentum Distillation](https://mlsanity.com/paper/2107.07651/) (2021)
- [CodeBERT: A Pre-Trained Model for Programming and Natural Languages](https://mlsanity.com/paper/2002.08155/) (2020)
- [Frozen in Time: A Joint Video and Image Encoder for End-to-End Retrieval](https://mlsanity.com/paper/2104.00650/) (2021)
- [CodeT5: Identifier-aware Unified Pre-trained Encoder-Decoder Models for Code Understanding and Generation](https://mlsanity.com/paper/2109.00859/) (2021)
- [CodeXGLUE: A Machine Learning Benchmark Dataset for Code Understanding and Generation](https://mlsanity.com/paper/2102.04664/) (2021)
- [CodeSearchNet Challenge: Evaluating the State of Semantic Code Search](https://mlsanity.com/paper/1909.09436/) (2019)
- [GraphCodeBERT: Pre-training Code Representations with Data Flow](https://mlsanity.com/paper/2009.08366/) (2020)
- [Unified Pre-training for Program Understanding and Generation](https://mlsanity.com/paper/2103.06333/) (2021)
- [IntelliCode Compose: Code Generation Using Transformer](https://mlsanity.com/paper/2005.08025/) (2020)
- [Thinking Fast and Slow: Efficient Text-to-Visual Retrieval with Transformers](https://mlsanity.com/paper/2103.16553/) (2021)
- [CoSQA: 20,000+ Web Queries for Code Search and Question Answering](https://mlsanity.com/paper/2105.13239/) (2021)
- [SynCoBERT: Syntax-Guided Multi-Modal Contrastive Pre-Training for Code Representation](https://mlsanity.com/paper/2108.04556/) (2021)
- [When Deep Learning Met Code Search](https://mlsanity.com/paper/1905.03813/) (2019)
- [In-IDE Code Generation from Natural Language: Promise and Challenges](https://mlsanity.com/paper/2101.11149/) (2021)

## Built on by (most cited first, 4 of 4)

- [CrossCodeBench: Benchmarking Cross-Task Generalization of Source Code Models](https://mlsanity.com/paper/2302.04030/) (2023)
- [CCT-Code: Cross-Consistency Training for Multilingual Clone Detection and Code Search](https://mlsanity.com/paper/2305.11626/) (2023)
- [Searching by Code: a New SearchBySnippet Dataset and SnippeR Retrieval Model for Searching by Code Snippets](https://mlsanity.com/paper/2305.11625/) (2023)
- [SynH-Rank: Quality-Aware Code Search via Diverse Data Synthesis and Hierarchical Ranking Training](https://mlsanity.com/paper/2607.17139/) (2026)

## Abstract (arXiv)

The goal of natural language semantic code search is to retrieve a semantically relevant code snippet from a fixed set of candidates using a natural language query. Existing approaches are neither effective nor efficient enough towards a practical semantic code search system. In this paper, we propose an efficient and accurate semantic code search framework with cascaded fast and slow models, in which a fast transformer encoder model is learned to optimize a scalable index for fast retrieval followed by learning a slow classification-based re-ranking model to improve the performance of the top K results from the fast retrieval. To further reduce the high memory cost of deploying two separate models in practice, we propose to jointly train the fast and slow model based on a single transformer encoder with shared parameters. The proposed cascaded approach is not only efficient and scalable, but also achieves state-of-the-art results with an average mean reciprocal ranking (MRR) score of 0.7795 (across 6 programming languages) as opposed to the previous state-of-the-art result of 0.713 MRR on the CodeSearchNet benchmark.