# Adaptive Endpointing with Deep Contextual Multi-armed Bandits

arXiv: https://arxiv.org/abs/2303.13407 · HTML: https://mlsanity.com/paper/2303.13407/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 27 references that are in the map.
- First posted on arXiv 23 Mar 2023.

## Builds on (most cited first, 12 of 12)

- [A Tutorial on Thompson Sampling](https://mlsanity.com/paper/1707.02038/) (2017)
- [Concrete Dropout](https://mlsanity.com/paper/1705.07832/) (2017)
- [Deep Bayesian Bandits Showdown: An Empirical Comparison of Bayesian Deep Networks for Thompson Sampling](https://mlsanity.com/paper/1802.09127/) (2018)
- [Neural Thompson Sampling](https://mlsanity.com/paper/2010.00827/) (2020)
- [Contextual Adapters for Personalized Speech Recognition in Neural Transducers](https://mlsanity.com/paper/2205.13660/) (2022)
- [Fast Contextual Adaptation with Neural Associative Memory for On-Device Personalized Speech Recognition](https://mlsanity.com/paper/2110.02220/) (2021)
- [E2E Segmenter: Joint Segmenting and Decoding for Long-Form ASR](https://mlsanity.com/paper/2204.10749/) (2022)
- [Deep Contextual Multi-armed Bandits](https://mlsanity.com/paper/1807.09809/) (2018)
- [Turn-Taking Prediction for Natural Conversational Speech](https://mlsanity.com/paper/2208.13321/) (2022)
- [Personal VAD 2.0: Optimizing Personal Voice Activity Detection for On-Device Speech Recognition](https://mlsanity.com/paper/2204.03793/) (2022)
- [Endpoint Detection for Streaming End-to-End Multi-talker ASR](https://mlsanity.com/paper/2201.09979/) (2022)
- [When can I Speak? Predicting initiation points for spoken dialogue agents](https://mlsanity.com/paper/2208.03812/) (2022)

## Built on by (most cited first, 1 of 1)

- [Nightjar: Dynamic Adaptive Speculative Decoding for Large Language Models Serving](https://mlsanity.com/paper/2512.22420/) (2025)

## Abstract (arXiv)

Current endpointing (EP) solutions learn in a supervised framework, which does not allow the model to incorporate feedback and improve in an online setting. Also, it is a common practice to utilize costly grid-search to find the best configuration for an endpointing model. In this paper, we aim to provide a solution for adaptive endpointing by proposing an efficient method for choosing an optimal endpointing configuration given utterance-level audio features in an online setting, while avoiding hyperparameter grid-search. Our method does not require ground truth labels, and only uses online learning from reward signals without requiring annotated labels. Specifically, we propose a deep contextual multi-armed bandit-based approach, which combines the representational power of neural networks with the action exploration behavior of Thompson modeling algorithms. We compare our approach to several baselines, and show that our deep bandit models also succeed in reducing early cutoff errors while maintaining low latency.