chikap421/catlvdm

[ICLR 2026 - ReALM-GEN] This repository accompanies the paper "Corruption-Aware Training of Latent Video Diffusion Models for Robust Text-to-Video Generation"

22
/ 100
Experimental

This project helps video creators, marketers, and researchers generate high-quality video content from text descriptions, even when the input text or underlying data might have imperfections or 'noise'. It takes a text prompt as input and produces a short video clip. This is especially useful for professionals who need reliable video generation from diverse or less-than-perfect source materials.

No commits in the last 6 months.

Use this if you need to reliably generate creative video content from text prompts, and you anticipate that your text inputs might be ambiguous, incomplete, or contain 'noise' from real-world data collection.

Not ideal if your primary concern is generating hyper-realistic video from perfectly clean, unambiguous prompts without any robustness considerations.

video-generation creative-content synthetic-media text-to-video robust-AI
Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 5 / 25
Maturity 15 / 25
Community 0 / 25

How are scores calculated?

Stars

10

Forks

Language

Python

License

MIT

Last pushed

Jul 11, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/chikap421/catlvdm"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.