all AI news
Learning to Watermark LLM-generated Text via Reinforcement Learning
March 19, 2024, 4:41 a.m. | Xiaojun Xu, Yuanshun Yao, Yang Liu
cs.LG updates on arXiv.org arxiv.org
Abstract: We study how to watermark LLM outputs, i.e. embedding algorithmically detectable signals into LLM-generated text to track misuse. Unlike the current mainstream methods that work with a fixed LLM, we expand the watermark design space by including the LLM tuning stage in the watermark pipeline. While prior works focus on token-level watermark that embeds signals into the output, we design a model-level watermark that embeds signals into the LLM weights, and such signals can be …
arxiv cs.ai cs.cr cs.lg generated llm reinforcement reinforcement learning text type via watermark
More from arxiv.org / cs.LG updates on arXiv.org
Sliced Wasserstein with Random-Path Projecting Directions
2 days, 20 hours ago |
arxiv.org
Learning Extrinsic Dexterity with Parameterized Manipulation Primitives
2 days, 20 hours ago |
arxiv.org
The Un-Kidnappable Robot: Acoustic Localization of Sneaking People
2 days, 20 hours ago |
arxiv.org
Jobs in AI, ML, Big Data
Data Engineer
@ Lemon.io | Remote: Europe, LATAM, Canada, UK, Asia, Oceania
Artificial Intelligence – Bioinformatic Expert
@ University of Texas Medical Branch | Galveston, TX
Lead Developer (AI)
@ Cere Network | San Francisco, US
Research Engineer
@ Allora Labs | Remote
Ecosystem Manager
@ Allora Labs | Remote
Founding AI Engineer, Agents
@ Occam AI | New York