# Benchmarking Commercial Intent Detection Services with Practice-Driven Evaluations

arXiv: https://arxiv.org/abs/2012.03929 · HTML: https://mlsanity.com/paper/2012.03929/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 7 Dec 2020.

## Built on by (most cited first, 8 of 8)

- [Improved Text Classification via Contrastive Adversarial Training](https://mlsanity.com/paper/2107.10137/) (2021)
- [Emergent Quantized Communication](https://mlsanity.com/paper/2211.02412/) (2022)
- [QAID: Question Answering Inspired Few-shot Intent Detection](https://mlsanity.com/paper/2303.01593/) (2023)
- [Z-BERT-A: a zero-shot Pipeline for Unknown Intent detection](https://mlsanity.com/paper/2208.07084/) (2022)
- [REIC: RAG-Enhanced Intent Classification at Scale](https://mlsanity.com/paper/2506.00210/) (2025)
- [Fast and Light-Weight Answer Text Retrieval in Dialogue Systems](https://mlsanity.com/paper/2205.14226/) (2022)
- [Practical token pruning for foundation models in few-shot conversational virtual assistant systems](https://mlsanity.com/paper/2408.11799/) (2024)
- [Natural Language Sentence Generation from API Specifications](https://mlsanity.com/paper/2206.06868/) (2022)

## Abstract (arXiv)

Intent detection is a key component of modern goal-oriented dialog systems that accomplish a user task by predicting the intent of users' text input. There are three primary challenges in designing robust and accurate intent detection models. First, typical intent detection models require a large amount of labeled data to achieve high accuracy. Unfortunately, in practical scenarios it is more common to find small, unbalanced, and noisy datasets. Secondly, even with large training data, the intent detection models can see a different distribution of test data when being deployed in the real world, leading to poor accuracy. Finally, a practical intent detection model must be computationally efficient in both training and single query inference so that it can be used continuously and re-trained frequently. We benchmark intent detection methods on a variety of datasets. Our results show that Watson Assistant's intent detection model outperforms other commercial solutions and is comparable to large pretrained language models while requiring only a fraction of computational resources and training data. Watson Assistant demonstrates a higher degree of robustness when the training and test distributions differ.