> For the complete documentation index, see [llms.txt](https://lynlabs.gitbook.io/lyn/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://lynlabs.gitbook.io/lyn/ai-modeling-research.md).

# AI Modeling Research

- [Autoregressive Modeling with Vector Quantization](https://lynlabs.gitbook.io/lyn/ai-modeling-research/autoregressive-modeling-with-vector-quantization.md): Features and approaches of the first-of-its-kind foundational video AI model powering video agents in the Lyn ecosystem.
- [Hierarchical Spatial-Temporal Video Generation Architecture](https://lynlabs.gitbook.io/lyn/ai-modeling-research/autoregressive-modeling-with-vector-quantization/hierarchical-spatial-temporal-video-generation-architecture.md): Encoding video into multi-scale latent video tokens and decoding video tokens back into the pixel 21 domain.
- [Efficient Autoregressive Video Generation via Token Masking](https://lynlabs.gitbook.io/lyn/ai-modeling-research/autoregressive-modeling-with-vector-quantization/efficient-autoregressive-video-generation-via-token-masking.md): Efficient modeling of complex spatial-temporal dynamics in video data.
- [Autoregressive Text-to-Visual Generation via Hybrid Architecture](https://lynlabs.gitbook.io/lyn/ai-modeling-research/autoregressive-modeling-with-vector-quantization/autoregressive-text-to-visual-generation-via-hybrid-architecture.md): A unique hybrid architecture of Mamba and Transformer for visual generation.
- [From Video to Movie: Composite Video Editing and RHF for Quality](https://lynlabs.gitbook.io/lyn/ai-modeling-research/from-video-to-movie-composite-video-editing-and-rhf-for-quality.md): A new framework extending from autoregressive video generation for industry-leading video edit precision, and RHF-based generation quality.
- [VideoGen-of-Thought](https://lynlabs.gitbook.io/lyn/ai-modeling-research/from-video-to-movie-composite-video-editing-and-rhf-for-quality/videogen-of-thought.md): A novel approach to the generation of long, consistently structured and homogenous content.
- [Supercharging MLLMs and LVLMs](https://lynlabs.gitbook.io/lyn/ai-modeling-research/supercharging-mllms-and-lvlms.md): Multi-modal Robustness benchmark (MMR) and Text-relevant Visual Token Selection (TVTS) developed for a better, open video AI.