UCSC-VLAA/Sight-Beyond-Text

[TMLR 2024] Official implementation of "Sight Beyond Text: Multi-Modal Training Enhances LLMs in Truthfulness and Ethics"

27
/ 100
Experimental

This project offers pre-trained components to make large language models (LLMs) more truthful and ethical by incorporating visual information. It takes existing LLMs and image datasets to produce an enhanced model that understands and processes both text and images more responsibly. AI researchers and developers working on safer, more reliable AI systems would use this.

No commits in the last 6 months.

Use this if you are a researcher or developer aiming to improve the trustworthiness and ethical behavior of your language models by adding multimodal capabilities.

Not ideal if you are looking for a ready-to-use, consumer-facing AI application, as this project provides research-focused components for model development.

AI ethics Multimodal AI Truthful AI Responsible AI development Large Language Models
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 6 / 25
Maturity 16 / 25
Community 5 / 25

How are scores calculated?

Stars

20

Forks

1

Language

Python

License

Apache-2.0

Last pushed

Sep 15, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/UCSC-VLAA/Sight-Beyond-Text"

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