# Efficient Temporal Sentence Grounding in Videos with Multi-Teacher Knowledge Distillation

arXiv: https://arxiv.org/abs/2308.03725 · HTML: https://mlsanity.com/paper/2308.03725/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in Multimodal distillation, one of 12 directions in Knowledge distillation.
- Builds on 21 of its 31 references that are in the map.
- First posted on arXiv 7 Aug 2023.

## Placed in

- Knowledge distillation / [Multimodal distillation](https://mlsanity.com/distill/multimodal-distillation/)

## Builds on (most cited first, 21 of 21)

- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset](https://mlsanity.com/paper/1705.07750/) (2017)
- [Visualizing and Understanding Convolutional Networks](https://mlsanity.com/paper/1311.2901/) (2013)
- [FitNets: Hints for Thin Deep Nets](https://mlsanity.com/paper/1412.6550/) (2014)
- [TALL: Temporal Activity Localization via Language Query](https://mlsanity.com/paper/1705.02101/) (2017)
- [Hollywood in Homes: Crowdsourcing Data Collection for Activity Understanding](https://mlsanity.com/paper/1604.01753/) (2016)
- [Knowledge Distillation and Student-Teacher Learning for Visual Intelligence: A Review and New Outlooks](https://mlsanity.com/paper/2004.05937/) (2020)
- [Learning 2D Temporal Adjacent Networks for Moment Localization with Natural Language](https://mlsanity.com/paper/1912.03590/) (2019)
- [Span-based Localizing Network for Natural Language Video Localization](https://mlsanity.com/paper/2004.13931/) (2020)
- [Knowledge Distillation by On-the-Fly Native Ensemble](https://mlsanity.com/paper/1806.04606/) (2018)
- [Semantic Conditioned Dynamic Modulation for Temporal Sentence Grounding in Videos](https://mlsanity.com/paper/1910.14303/) (2019)
- [Boundary Proposal Network for Two-Stage Natural Language Video Localization](https://mlsanity.com/paper/2103.08109/) (2021)
- [Temporally Grounding Language Queries in Videos by Contextual Boundary-aware Prediction](https://mlsanity.com/paper/1909.05010/) (2019)
- [Online Knowledge Distillation for Efficient Pose Estimation](https://mlsanity.com/paper/2108.02092/) (2021)
- [Temporal Sentence Grounding in Videos: A Survey and Future Directions](https://mlsanity.com/paper/2201.08071/) (2022)
- [Exploring Motion and Appearance Information for Temporal Sentence Grounding](https://mlsanity.com/paper/2201.00457/) (2022)
- [Video Activity Localisation with Uncertainties in Temporal Boundary](https://mlsanity.com/paper/2206.12923/) (2022)
- [Parallel Attention Network with Sequence Matching for Video Grounding](https://mlsanity.com/paper/2105.08481/) (2021)
- [Meta Knowledge Distillation](https://mlsanity.com/paper/2202.07940/) (2022)
- [Entity-aware and Motion-aware Transformers for Language-driven Action Localization in Videos](https://mlsanity.com/paper/2205.05854/) (2022)
- [Learning Commonsense-aware Moment-Text Alignment for Fast Video Temporal Grounding](https://mlsanity.com/paper/2204.01450/) (2022)

## Built on by (most cited first, 3 of 3)

- [MambaTAD: When State-Space Models Meet Long-Range Temporal Action Detection](https://mlsanity.com/paper/2511.17929/) (2025)
- [Video sentence grounding with temporally global textual knowledge](https://mlsanity.com/paper/2404.13611/) (2024)
- [CertVLA: Certified Defense against Physical Visual Attacks for Vision-Language-Action Models](https://mlsanity.com/paper/2608.20791/) (2026)

## Abstract (arXiv)

Temporal Sentence Grounding in Videos (TSGV) aims to detect the event timestamps described by the natural language query from untrimmed videos. This paper discusses the challenge of achieving efficient computation in TSGV models while maintaining high performance. Most existing approaches exquisitely design complex architectures to improve accuracy with extra layers and loss, suffering from inefficiency and heaviness. Although some works have noticed that, they only make an issue of feature fusion layers, which can hardly enjoy the highspeed merit in the whole clunky network. To tackle this problem, we propose a novel efficient multi-teacher model (EMTM) based on knowledge distillation to transfer diverse knowledge from both heterogeneous and isomorphic networks. Specifically, We first unify different outputs of the heterogeneous models into one single form. Next, a Knowledge Aggregation Unit (KAU) is built to acquire high-quality integrated soft labels from multiple teachers. After that, the KAU module leverages the multi-scale video and global query information to adaptively determine the weights of different teachers. A Shared Encoder strategy is then proposed to solve the problem that the student shallow layers hardly benefit from teachers, in which an isomorphic teacher is collaboratively trained with the student to align their hidden states. Extensive experimental results on three popular TSGV benchmarks demonstrate that our method is both effective and efficient without bells and whistles.