Authors:
(1) Dan Kondratyuk, Google Research and with Equal contribution;
(2) Lijun Yu, Google Research, Carnegie Mellon University and with Equal contribution;
(3) Xiuye Gu, Google Research and with Equal contribution;
(4) Jose Lezama, Google Research and with Equal contribution;
(5) Jonathan Huang, Google Research and with Equal contribution;
(6) Grant Schindler, Google Research;
(7) Rachel Hornung, Google Research;
(8) Vighnesh Birodkar, Google Research;
(9) Jimmy Yan, Google Research;
(10) Krishna Somandepalli, Google Research;
(11) Hassan Akbari, Google Research;
(12) Yair Alon, Google Research;
(13) Yong Cheng, Google DeepMind;
(14) Josh Dillon, Google Research;
(15) Agrim Gupta, Google Research;
(16) Meera Hahn, Google Research;
(17) Anja Hauth, Google Research;
(18) David Hendon, Google Research;
(19) Alonso Martinez, Google Research;
(20) David Minnen, Google Research;
(21) Mikhail Sirotenko, Google Research;
(22) Kihyuk Sohn, Google Research;
(23) Xuan Yang, Google Research;
(24) Hartwig Adam, Google Research;
(25) Ming-Hsuan Yang, Google Research;
(26) Irfan Essa, Google Research;
(27) Huisheng Wang, Google Research;
(28) David A. Ross, Google Research;
(29) Bryan Seybold, Google Research and with Equal contribution;
(30) Lu Jiang, Google Research and with Equal contribution. Table of Links Abstract and 1 Introduction 2. Related Work 3. Model Overview and 3.1. Tokenization 3.2. Language Model Backbone and 3.3. Super-Resolution 4. LLM Pretraining for Generation 4.1. Task Prompt Design 4.2. Training Strategy 5. Experiments 5.1. Experimental Setup 5.2. Pretraining Task Analysis 5.3. Comparison with the State-of-the-Art 5.4. LLM’s Diverse Capabilities in Video Generation and 5.5. Limitations 6. Conclusion, Acknowledgements, and References A. Appendix 5.1. Experimental Setup Training tasks. We train the model on a mixture of pretraining tasks as detailed in Section 4.1. We finetune a model on a high-quality training subset for text-to-video evaluations, as discussed in Section 4.2. Unless explicitly stated, we do not finetune on specific tasks for evaluations. Datasets. We train on a total of 1B image-text pairs and ∼270M videos (∼100M with paired text, of which ∼50M are used for high-quality finetuning, and ∼170M with paired audio) from the public internet and other sources, i.e. around 2 trillion tokens across all modalities. The data has been filtered to remove egregious content and sampled to improve contextual and demographic diversity. Evaluation protocol. We employ a zero-shot generation evaluation protocol, as the model has not been trained on the training data of target benchmarks. Specifically, the evaluation benchmark includes two text-to-video generation datasets, MSR-VTT (Xu et al., 2016) and UCF-101 (Soomro et al., 2012), as well as the frame prediction task on Kinetics 600 (K600) (Carreira et al., 2018), in which the first 5 frames are provided as the condition to predict the next 11 frames. We also include inpainting and outpainting tasks (Yu et al., 2023a) on Something-Something V2 (SSv2) (Goyal et al., 2017). We adopt widely used metrics such as Frechet Video Distance (FVD) (Unterthiner et al., 2018), CLIP similarity score (Wu et al., 2021), and Inception Score (IS) (Saito et al., 2020) for evaluation. Note that the specific metrics and evaluation methods vary across different datasets. Detailed information on these variations can be found in Appendix A.5.4. We include examples of the generated videos in the supplementary materials. This paper is available on arxiv under CC BY 4.0 DEED license. Authors: (1) Dan Kondratyuk, Google Research and with Equal contribution; (2) Lijun Yu, Google Research, Carnegie Mellon University and with Equal contribution; (3) Xiuye Gu, Google Research and with Equal contribution; (4) Jose Lezama, Google Research and with Equal contribution; (5) Jonathan Huang, Google Research and with Equal contribution; (6) Grant Schindler, Google Research; (7) Rachel Hornung, Google Research; (8) Vighnesh Birodkar, Google Research; (9) Jimmy Yan, Google Research; (10) Krishna Somandepalli, Google Research; (11) Hassan Akbari, Google Research; (12) Yair Alon, Google Research; (13) Yong Cheng, Google DeepMind; (14) Josh Dillon, Google Research; (15) Agrim Gupta, Google Research; (16) Meera Hahn, Google Research; (17) Anja Hauth, Google Research; (18) David Hendon, Google Research; (19) Alonso Martinez, Google Research; (20) David Minnen, Google Research; (21) Mikhail Sirotenko, Google Research; (22) Kihyuk Sohn, Google Research; (23) Xuan Yang, Google Research; (24) Hartwig Adam, Google Research; (25) Ming-Hsuan Yang, Google Research; (26) Irfan Essa, Google Research; (27) Huisheng Wang, Google Research; (28) David A. Ross, Google Research; (29) Bryan Seybold, Google Research and with Equal contribution; (30) Lu Jiang, Google Research and with Equal contribution. Authors: Authors: (1) Dan Kondratyuk, Google Research and with Equal contribution; (2) Lijun Yu, Google Research, Carnegie Mellon University and with Equal contribution; (3) Xiuye Gu, Google Research and with Equal contribution; (4) Jose Lezama, Google Research and with Equal contribution; (5) Jonathan Huang, Google Research and with Equal contribution; (6) Grant Schindler, Google Research; (7) Rachel Hornung, Google Research; (8) Vighnesh Birodkar, Google Research; (9) Jimmy Yan, Google Research; (10) Krishna Somandepalli, Google Research; (11) Hassan Akbari, Google Research; (12) Yair Alon, Google Research; (13) Yong Cheng, Google DeepMind; (14) Josh Dillon, Google Research; (15) Agrim Gupta, Google Research; (16) Meera Hahn, Google Research; (17) Anja Hauth, Google Research; (18) David Hendon, Google Research; (19) Alonso Martinez, Google Research; (20) David Minnen, Google Research; (21) Mikhail Sirotenko, Google Research; (22) Kihyuk Sohn, Google Research; (23) Xuan Yang, Google Research; (24) Hartwig Adam, Google Research; (25) Ming-Hsuan Yang, Google Research; (26) Irfan Essa, Google Research; (27) Huisheng Wang, Google Research; (28) David A. Ross, Google Research; (29) Bryan Seybold, Google Research and with Equal contribution; (30) Lu Jiang, Google Research and with Equal contribution. Table of Links Abstract and 1 Introduction Abstract and 1 Introduction 2. Related Work 2. Related Work 3. Model Overview and 3.1. Tokenization 3. Model Overview and 3.1. Tokenization 3.2. Language Model Backbone and 3.3. Super-Resolution 3.2. Language Model Backbone and 3.3. Super-Resolution 4. LLM Pretraining for Generation 4.1. Task Prompt Design 4.1. Task Prompt Design 4.2. Training Strategy 4.2. Training Strategy 5. Experiments 5.1. Experimental Setup 5.1. Experimental Setup 5.2. Pretraining Task Analysis 5.2. Pretraining Task Analysis 5.3. Comparison with the State-of-the-Art 5.3. Comparison with the State-of-the-Art 5.4. LLM’s Diverse Capabilities in Video Generation and 5.5. Limitations 5.4. LLM’s Diverse Capabilities in Video Generation and 5.5. Limitations 6. Conclusion, Acknowledgements, and References 6. Conclusion, Acknowledgements, and References A. Appendix A. Appendix 5.1. Experimental Setup Training tasks. We train the model on a mixture of pretraining tasks as detailed in Section 4.1. We finetune a model on a high-quality training subset for text-to-video evaluations, as discussed in Section 4.2. Unless explicitly stated, we do not finetune on specific tasks for evaluations. Training tasks. Datasets. We train on a total of 1B image-text pairs and ∼270M videos (∼100M with paired text, of which ∼50M are used for high-quality finetuning, and ∼170M with paired audio) from the public internet and other sources, i.e. around 2 trillion tokens across all modalities. The data has been filtered to remove egregious content and sampled to improve contextual and demographic diversity. Datasets. Evaluation protocol. We employ a zero-shot generation evaluation protocol, as the model has not been trained on the training data of target benchmarks. Specifically, the evaluation benchmark includes two text-to-video generation datasets, MSR-VTT (Xu et al., 2016) and UCF-101 (Soomro et al., 2012), as well as the frame prediction task on Kinetics 600 (K600) (Carreira et al., 2018), in which the first 5 frames are provided as the condition to predict the next 11 frames. We also include inpainting and outpainting tasks (Yu et al., 2023a) on Something-Something V2 (SSv2) (Goyal et al., 2017). Evaluation protocol. We adopt widely used metrics such as Frechet Video Distance (FVD) (Unterthiner et al., 2018), CLIP similarity score (Wu et al., 2021), and Inception Score (IS) (Saito et al., 2020) for evaluation. Note that the specific metrics and evaluation methods vary across different datasets. Detailed information on these variations can be found in Appendix A.5.4. We include examples of the generated videos in the supplementary materials. This paper is available on arxiv under CC BY 4.0 DEED license. This paper is available on arxiv under CC BY 4.0 DEED license. available on arxiv

Part of HackerNoon's growing list of open-source research papers, promoting free access to academic material.

Pretraining Task Analysis On LLM Video Generati

Task Prompt Design For LLM Video Generation

Teleplay.Tech

Too Long; Didn't Read

Make resilience your competitive advantage

Experimental Setup For Large Language Model Video Generation

About Author

Comments

TOPICS

THIS ARTICLE WAS FEATURED IN

Related Stories

Untitled Story

Advancing Coherent Story Generation: Future Directions for AI Writing Tools

100 Days of AI Day 2: Enhancing Prompt Engineering for ChatGPT

The Noonification: Exploring Tool-Integrated Reasoning: Innovating Math-Proficient LLMs (10/5/2023)

The Noonification: Simple Database Migration Scripts On Your CI/CD step (10/16/2023)

The Noonification: Breaking Axioms in Program Execution (11/5/2023)

Advancing Coherent Story Generation: Future Directions for AI Writing Tools

100 Days of AI Day 2: Enhancing Prompt Engineering for ChatGPT

The Noonification: Exploring Tool-Integrated Reasoning: Innovating Math-Proficient LLMs (10/5/2023)

The Noonification: Simple Database Migration Scripts On Your CI/CD step (10/16/2023)

The Noonification: Breaking Axioms in Program Execution (11/5/2023)

Light-Mode

Classic

Newspaper

Dark-Mode

Neon Noir

Minty

HN StartUps