Memory-Based Model Editing at Scale

1 code implementation • 13 Jun 2022 • Eric Mitchell, Charles Lin, Antoine Bosselut, Christopher D. Manning, Chelsea Finn

We find that only SERAC achieves high performance on all three problems, consistently outperforming existing approaches to model editing by a significant margin.

counterfactual Dialogue Generation +5

Fast Model Editing at Scale

3 code implementations • ICLR 2022 • Eric Mitchell, Charles Lin, Antoine Bosselut, Chelsea Finn, Christopher D. Manning

To enable easy post-hoc editing at scale, we propose Model Editor Networks using Gradient Decomposition (MEND), a collection of small auxiliary editing networks that use a single desired input-output pair to make fast, local edits to a pre-trained model's behavior.

Language Modelling Model Editing

LASER: Learning a Latent Action Space for Efficient Reinforcement Learning

no code implementations • 29 Mar 2021 • Arthur Allshire, Roberto Martín-Martín, Charles Lin, Shawn Manuel, Silvio Savarese, Animesh Garg

Additionally, similar tasks or instances of the same task family impose latent manifold constraints on the most effective action space: the task family can be best solved with actions in a manifold of the entire action space of the robot.

reinforcement-learning Reinforcement Learning (RL)

The Stanford Acuity Test: A Precise Vision Test Using Bayesian Techniques and a Discovery in Human Visual Response

no code implementations • 5 Jun 2019 • Chris Piech, Ali Malik, Laura M Scott, Robert T. Chang, Charles Lin

First, we uncover a new parametric probabilistic model of visual acuity response based on detailed measurements of patients with eye disease.