Language Model AND Sanjeev Arora
Common descendants
7 Documents
[2307.15936] A Theory for Emergence of Complex Skills in Language Models
Tags:
2024-02-24 About
Sanjeev Arora sur X : "It's better to use just 5% of the instruction-tuning data (suitably selected) instead of the full dataset."
Tags:
2024-02-15 About
New Theory Suggests Chatbots Can Understand Text | Quanta Magazine
Tags:
2024-02-11 About
[2302.06600] Task-Specific Skill Localization in Fine-tuned Language Models
Tags:
2023-08-25 About
Sanjeev Arora sur Twitter : "new `skills' induced by LLM fine-tuning can be localized in tiny fraction of the model."
Tags:
2023-07-07 About
Sanjeev Arora sur Twitter : "Fine-tuning language models using just forward pass!...r
Tags:
2023-06-09 About
Sanjeev Arora sur Twitter : "A priori, fine-tuning a huge LM on a few datapoints could lead to catastrophic overfitting. So why doesn’t it? Our theory + experiments..."
Tags:
2022-10-14 About