Skip to content
Trending storyDeveloping

Ledell Wu explains human-calibrated evaluation for improving video generation

1 article1 sourcesince Nov 5Last article 1h ago ·

Overview

AISummary of 1 article

Ledell Wu, speaking as part of Creatify AI's Boreal work on self-improving video generation, explains how human-calibrated evaluation guides improvements to data, training, and inference for video models.

He says the feedback helps video models learn. The report, posted by LMSYS Org on X, gives no figures, benchmarks, or released results.

Written by AI from the articles below · updated Nov 5, 5:54 PM ET

Check the sources:

Latest developmentNov 5, 5:43 PM ET
Ledell Wu explains how human-calibrated evaluation improves video generation

Article timeline

The articles in this story. Times are ET.

Nov 5
  1. LMSYS OrgOfficial
    Ledell Wu explains how human-calibrated evaluation improves video generation

    AILedell Wu shares how human-calibrated evaluation guides improvements to data, training, and inference for video models. He says the feedback helps video models learn, as part of Creatify AI's Boreal work on self-improving video generation.

    Image from @lmsysorg's post

Heat Trend

Not enough continuous observations to show a trend yet.