# Exploring Vision-Language Models for Open-Vocabulary Zero-Shot Action Segmentation

arXiv: https://arxiv.org/abs/2602.21406 · HTML: https://mlsanity.com/paper/2602.21406/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Video understanding with MLLMs, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 24 of its 35 references that are in the map.
- First posted on arXiv 24 Feb 2026.

## Placed in

- Multimodal LLMs / vision-language models / [Video understanding with MLLMs](https://mlsanity.com/mllm/video-understanding-with-mllms/)

## Builds on (most cited first, 24 of 24)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Sigmoid Loss for Language Image Pre-Training](https://mlsanity.com/paper/2303.15343/) (2023)
- [Perception Encoder: The best visual embeddings are not at the output of the network](https://mlsanity.com/paper/2504.13181/) (2025)
- [Prompting Visual-Language Models for Efficient Video Understanding](https://mlsanity.com/paper/2112.04478/) (2021)
- [Alleviating Over-segmentation Errors by Detecting Action Boundaries](https://mlsanity.com/paper/2007.06866/) (2020)
- [Temporal Action Segmentation: An Analysis of Modern Techniques](https://mlsanity.com/paper/2210.10352/) (2022)
- [Weakly Supervised Action Learning with RNN based Fine-to-coarse Modeling](https://mlsanity.com/paper/1703.08132/) (2017)
- [NeuralNetwork-Viterbi: A Framework for Weakly Supervised Video Learning](https://mlsanity.com/paper/1805.06875/) (2018)
- [Temporally-Weighted Hierarchical Clustering for Unsupervised Action Segmentation](https://mlsanity.com/paper/2103.11264/) (2021)
- [Temporal Action Segmentation from Timestamp Supervision](https://mlsanity.com/paper/2103.06669/) (2021)
- [Weakly supervised learning of actions from transcripts](https://mlsanity.com/paper/1610.02237/) (2016)
- [Temporally Consistent Unbalanced Optimal Transport for Unsupervised Action Segmentation](https://mlsanity.com/paper/2404.01518/) (2024)
- [Set-Constrained Viterbi for Set-Supervised Action Segmentation](https://mlsanity.com/paper/2002.11925/) (2020)
- [Action Shuffle Alternating Learning for Unsupervised Action Segmentation](https://mlsanity.com/paper/2104.02116/) (2021)
- [Unsupervised Discriminative Embedding for Sub-Action Learning in Complex Activities](https://mlsanity.com/paper/2105.00067/) (2021)
- [Hierarchical Vector Quantization for Unsupervised Action Segmentation](https://mlsanity.com/paper/2412.17640/) (2024)
- [Timestamp-Supervised Action Segmentation with Graph Convolutional Networks](https://mlsanity.com/paper/2206.15031/) (2022)
- [Leveraging Action Affinity and Continuity for Semi-supervised Temporal Action Segmentation](https://mlsanity.com/paper/2207.08653/) (2022)
- [SSCAP: Self-supervised Co-occurrence Action Parsing for Unsupervised Temporal Action Segmentation](https://mlsanity.com/paper/2105.14158/) (2021)
- [Hierarchical NeuroSymbolic Approach for Comprehensive and Explainable Action Quality Assessment](https://mlsanity.com/paper/2403.13798/) (2024)
- [Fine-grained Action Segmentation using the Semi-Supervised Action GAN](https://mlsanity.com/paper/1909.09269/) (2019)
- [Leveraging triplet loss for unsupervised action segmentation](https://mlsanity.com/paper/2304.06403/) (2023)
- [Building a Multi-modal Spatiotemporal Expert for Zero-shot Action Recognition with CLIP](https://mlsanity.com/paper/2412.09895/) (2024)
- [A Backbone for Long-Horizon Robot Task Understanding](https://mlsanity.com/paper/2408.01334/) (2024)

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

- [The Unreasonable Effectiveness of VLMs for Zero-shot Procedural Mistake Detection](https://mlsanity.com/paper/2606.21579/) (2026)

## Abstract (arXiv)

Temporal Action Segmentation (TAS) requires dividing videos into action segments, yet the vast space of activities and alternative breakdowns makes collecting comprehensive datasets infeasible. Existing methods remain limited to closed vocabularies and fixed label sets. In this work, we explore the largely unexplored problem of Open-Vocabulary Zero-Shot Temporal Action Segmentation (OVTAS) by leveraging the strong zero-shot capabilities of Vision-Language Models (VLMs). We introduce a training-free pipeline that follows a segmentation-by-classification design: Frame-Action Embedding Similarity (FAES) matches video frames to candidate action labels, and Similarity-Matrix Temporal Segmentation (SMTS) enforces temporal consistency. Beyond proposing OVTAS, we present a systematic study across 14 diverse VLMs, providing the first broad analysis of their suitability for open-vocabulary action segmentation. Experiments on standard benchmarks show that OVTAS achieves strong results without task-specific supervision, underscoring the potential of VLMs for structured temporal understanding.