6 accepted papers
Scaling the capacity of language models has consistently proven to be a reliable approach for improving performance and unlocking new capabilities. Capacity can be primarily defined by two dimensions: the number of model parameters and the compute per example. While scaling typically involves increa…
How do language models use information provided as context when generating a response? Can we infer whether a particular generated statement is actually grounded in the context, a misinterpretation, or fabricated? To help answer these questions, we introduce the problem of *context attribution*: pin…
*How does the internal computation of a machine learning model transform inputs into predictions?* To tackle this question, we introduce a framework called *component modeling* for decomposing a model prediction in terms of its components---architectural "building blocks" such as convolution filters…
We study the problem of (learning) algorithm comparison, where the goal is to find differences between models trained with two different learning algorithms. We begin by formalizing this goal as one of finding distinguishing feature transformations, i.e., input transformations that change the predic…
Post-hoc gradient-based interpretability methods [Simonyan et al., 2013, Smilkov et al., 2017] that provide instance-specific explanations of model predictions are often based on assumption (A): magnitude of input gradients—gradients of logits with respect to input—noisily highlight discriminative t…
Several works have proposed Simplicity Bias (SB)---the tendency of standard training procedures such as Stochastic Gradient Descent (SGD) to find simple models---to justify why neural networks generalize well [Arpit et al. 2017, Nakkiran et al. 2019, Valle-Perez et al. 2019]. However, the precise no…