Alice Schoenauer Sebag

Are you curious about the implications of using LLMs for hiring? 🤔 The safety team at Cohere is studying the fairness of LLMs in real-world use cases, and you can help out the investigation by sharing your resume here 👉 forms.gle/4YFA3ztBc6rP5g…

@lio_train_sncf Bonjour, le train TER 870160 qui doit partir demain à 16:18 de Figeac pour Brive circulera-t-il ou est-il remplacé par un bus ? Si oui à quelle heure partira le bus ? Merci !

🚦 #FlashInfoliOTrain 7h30 Le trafic est conforme aux prévisions, les #liOTrain sillonnent notre belle région #Occitanie en temps et en heure ce matin ☑️🌄 Nous restons attentifs à l'évolution des circulations jusqu'à la fin de cette session, prévue à 20h 🙋🏻‍♂️ Une ❓ 🟰 💬👇🏻📲

Announcing C4AI Command R+ open weights, a state-of-the-art 104B LLM with RAG, tooling and multilingual in 10 languages.  This release builds on our 35B and is a part of our commitment to make AI breakthroughs accessible to the research community. 🎉 huggingface.co/CohereForAI/c4…

« Prévenir les cancers implique de taxer, limiter et surtout interdire des substances aussi pathogènes que rentables » lemonde.fr/idees/article/…

Could it be that this is *not* doing ourselves a favor, by giving some independence of thought to something we’re fully responsible for? 3/3 🙊 #ai #airegulation #confusion #paradigm #wordsmatter

But data management infra/soft don’t have a soul afaik 😅 they are human-designed, executed, tested and deployed. Could it be that we’re losing track of this, like we’re doing when we call « artificial intelligences » systems that are human-designed, trained, tested and deployed?

Have you also noticed ppl increasingly saying « system confusion » for « bug »? At the post office, I was recently told that xyz would « really confuse the system ». At the airport, I had issues at check in because my « name is too long, and it confused the system » (!) 1/n

« How to talk to a science denier » by @LeeCMcIntyre : a rigorous examination, both from a personal and academic points of view, of what is and what can be done when faced with, science denial. Definitely recommend! mitpress.mit.edu/9780262545051/…

A 🧵worth reading on a hot topic

🏭💧After months of work, the Forever Pollution Project by @lemondefr and its 17 partners reveals the staggering extent of Europe's contamination by forever chemicals #PFAS #ForeverPollution #Crossborder lemonde.fr/en/les-decodeu…

Language models (LMs) exhibit harmful biases that can get worse with size. Reinforcement learning from human feedback (RLHF) helps, but not always enough. We show that simple prompting approaches can help LMs trained with RLHF produce less harmful outputs. arxiv.org/abs/2302.07459

⬇️😉

@matthiaseck Enjoy!!

SFO -> FRA

A clip from Don’t Look Up, and then a real TV interview that just happened

450+ sister marches across the country. 1 MILLION protestors nationwide – including 40k in Washington, D.C. Our movement is strong, angry, and ready to fight for abortion access.

Fuck around & find out. It’s time to mobilize. @womensmarch is mobilizing on 5/14 with @PPFA, @UltraViolet, and @MoveOn to tell the Supreme Court that we’re going to fucking fight for abortion rights. Get in the streets with us: act.womensmarch.com/sign/bans-off-…

"This will fall on the poorest women [...] on young women who have been abused, who are victims of incest [...] on those who have been raped [...] on mothers who are already struggling to work 3 jobs to be able to support the children they have. [...] We are not going back"

@NowImCassie Sending hugs 💝 🫂

Going on private for a bit cuz transphobic Apartheid boy stans found my Tweets and started harassing me