Jonas Geiping's picture

8 9 11

Jonas Geiping

JonasGeiping

·

https://jonasgeiping.github.io/

AI & ML interests

Machine Learning Safety, Security and Privacy; Optimization in Deep Learning; Mathematical Optimization: Federated Learning

Organizations

JonasGeiping's activity

upvoted a collection 3 months ago

Goldfish Loss: Mitigating Memorization in LLMs

This collection contains artifacts from our paper titled: "Be like a Goldfish, Don't Memorize! Mitigating Memorization in Generative LLMs." • 9 items • Updated 10 days ago • 2

upvoted 2 papers 5 months ago

Be like a Goldfish, Don't Memorize! Mitigating Memorization in Generative LLMs

Paper • 2406.10209 • Published Jun 14 • 8

From Pixels to Prose: A Large Dataset of Dense Image Captions

Paper • 2406.10328 • Published Jun 14 • 17

upvoted a paper 6 months ago

Transformers Can Do Arithmetic with the Right Embeddings

Paper • 2405.17399 • Published May 27 • 51

upvoted 2 papers 9 months ago

NEFTune: Noisy Embeddings Improve Instruction Finetuning

Paper • 2310.05914 • Published Oct 9, 2023 • 14

Coercing LLMs to do and reveal (almost) anything

Paper • 2402.14020 • Published Feb 21 • 12

upvoted a paper 10 months ago

Spotting LLMs With Binoculars: Zero-Shot Detection of Machine-Generated Text

Paper • 2401.12070 • Published Jan 22 • 42

upvoted a paper 11 months ago

Object Recognition as Next Token Prediction

Paper • 2312.02142 • Published Dec 4, 2023 • 11

upvoted a paper over 1 year ago

Does Circuit Analysis Interpretability Scale? Evidence from Multiple Choice Capabilities in Chinchilla

Paper • 2307.09458 • Published Jul 18, 2023 • 10