Abhimanyu Hans

2 papers · 2024–2024 · 2 conferences · across top CS/AI conferences

Achievements

🌉 Interdisciplinary Bridge 🌍 Conference Polyglot (2) 🐝 Cross-Pollinator (12) 🌈 Renaissance Researcher (5) 🧭 Keyword Pioneer

Conferences

ICML (1) NIPS (1)

Top co-authors

Hamid Kazemi (2) Tom Goldstein (2) Jonas Geiping (2) Avi Schwarzschild (1) Yuxin Wen (1) Aniruddha Saha (1) Gowthami Somepalli (1) Micah Goldblum (1) Valeriia Cherepanova (1) Prajwal Singhania (1)

Research topics

Keywords

generative model (1) next-token prediction (1) copyright protection (1) privacy protection (1) memorization mitigation (1) privacy risk (1) training objective (1) large language model (1) token dropping (1) verbatim reproduction (1)

Papers

Be like a Goldfish, Don't Memorize! Mitigating Memorization in Generative LLMs NIPS 2024

Spotting LLMs With Binoculars: Zero-Shot Detection of Machine-Generated Text ICML 2024