2/ Fact-checking with LLMs – investigates fact-checking capabilities of LLMs; shows the enhanced prowess of LLMs when equipped with contextual information; GPT4 outperforms GPT-3, but accuracy varies based on query language & claim veracity.
GENERATIVE AI
-
Robust Prompts Maintain Value Across LLM Variations
By
–
Yes. But some prompts are quite robust to noise (and to differences in LLMs), and that's their value.
-
Spooky Black Light SDXL Fine-Tune Model Released
By
–
Spooky black light SDXL fine tune anyone? https://
replicate.com/fofr/sdxl-blac
k-light
… -
AI Threatens Translation, Interpretation, and Film Dubbing Jobs
By
–
Vous avais-je déjà dit que je parlais toute les langues ?
— Rafik Smati (@RafikSmati) 29 octobre 2023
Blague à part : parmi les nombreux métiers qui vont disparaître à cause de l'IA, il y a les interprètes, les traducteurs et les doubleurs de films. En voici la preuve ⤵️ pic.twitter.com/xTFdurS51MVous avais-je déjà dit que je parlais toute les langues ? Blague à part : parmi les nombreux métiers qui vont disparaître à cause de l'IA, il y a les interprètes, les traducteurs et les doubleurs de films. En voici la preuve
-
DoReMi: Optimizing Data Mixtures for Faster Language Model Pretraining
By
–
Check out DoReMi: Optimizing Data Mixtures Speeds Up Language Model Pretraining, and some of the papers it cites. By @sangmichaelxie et al.
-
Pop culture milestone achieved with AI leaders Mitchell Gebru McMillan
By
–
Well, this is some kind of pop culture achievement unlocked. W/
@mmitchell_ai @timnitGebru @mcmillan_majora -
Character AI Discord Server Reaches 50K Members
By
–
Happy 50K on Discord!! http://
discord.gg/characterai -
Pika Labs Contest Prizes: Cash, NFT Access, and Social Promotion
By
–
Prizes: 1. 1000 Dollars cash prize for 1st place. 2. Surprise prizes for the 2nd, 3rd, and the most liked post on social media. (TBD)
3. 1-month access to non-watermark generations
4. Promotion of your work and social media accounts across all Pika official platforms.
