🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

Deduplicating Training Data Mitigates Privacy Risks in Language Models Finds that seqs that are present in training data tend to be super-linearly (extremely) over-represented in generated seqs and proposes a method to mitigate this issue. arxiv.org/abs/2202.06539

Music Signal Enhancement at #ICASSP2022! GANs + Denoising Diffusion to make noisy music signals sound cleaner and crispier. Our pre-print: arxiv.org/abs/2204.13289 Source code and more demos: nkandpa2.github.io/music-enhancem… Work with serious titans: Nikhil Kandpal and zeyu jin 🧵

Large Language Models Struggle to Learn Long-Tail Knowledge abs: arxiv.org/abs/2211.08411

Models such as Stable Diffusion are trained on copyrighted, trademarked, private, and sensitive images. Yet, our new paper shows that diffusion models memorize images from their training data and emit them at generation time. Paper: arxiv.org/abs/2301.13188 👇[1/9]

Is Kevin onto something? We found that LLMs can struggle to understand compressed text, unless you do some specific tricks. Check out arxiv.org/abs/2404.03626 and help Jaehoon Lee, Alex Alemi, Jeffrey Pennington, Adam Roberts, Jascha Sohl-Dickstein, Noah Constant and I make Kevin’s dream a reality.