Mastodon discussion Discussions Sep 8

Vision-language models are increasingly used as reward functions for robotic learning. Paraphrase-induced instability is...

by Sipirtu

Vision-language models are increasingly used as reward functions for robotic learning. Paraphrase-induced instability is widespread and severe across proprietary and open-source models.Source: arXiv cs.ROhttps://arxiv.org/abs/2609.05401#MachineLearning

Read Original
Multimodal

Metadata

Account
sipirtu

Related