Timo Schick (@timo_schick) / X

Timo Schick on X: 🎉New paper🎉 In Self-Diagnosis and Self-Debiasing, we investigate whether pretrained LMs can use their internal knowledge to discard undesired behaviors and reduce biases in their own outputs (w/@4digitaldignity + @

Timo Schick on X: PET ( now runs with the latest version of @huggingface's transformers library. This means it is now possible to perform zero-shot and few-shot PET learning with multilingual models

Michele Bevilacqua (@MicheleBevila20) / X

Timo Schick – Member of Technical Staff – Inflection AI

Timo Schick

Timo Schick on X: 🎉 New paper 🎉 We show that prompt-based learners like PET excel in true few-shot settings (@EthanJPerez) if correctly configured: On @oughtinc's RAFT, PET performs close to non-expert

Edouard Grave ✈️ NeurIPS 2023 (@EXGRV) / X

Timo Schick (@timo_schick) / X

Timo Schick on X: Interested in distilling zero-shot knowledge from big LMs like GPT-3? Or in learning more about a movie called Bullfrogs on Poopy Mountain? 🐸💩 Check out our blog post

Timo Schick (@timo_schick) / X

Emanuele Vivoli (@EmanueleVivoli) / X