Video-LLaVA creates a unified visual representation for images and videos via pre-projection alignment, enabling mutual enhancement from joint training and strong results on image and video benchmarks.
Title resolution pending
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2023 3representative citing papers
Hidden activations in LLMs encode detectable information about statement truthfulness, enabling a classifier to identify true versus false content more reliably than the model's assigned probabilities.
Baichuan 2 presents 7B and 13B LLMs trained on 2.6T tokens that match or exceed similar open models on MMLU, CMMLU, GSM8K, HumanEval and excel in medicine and law.
citing papers explorer
-
The Internal State of an LLM Knows When It's Lying
Hidden activations in LLMs encode detectable information about statement truthfulness, enabling a classifier to identify true versus false content more reliably than the model's assigned probabilities.