7 accepted papers
We present Direct Reward Fine-Tuning (DRaFT), a simple and effective method for fine-tuning diffusion models to maximize differentiable reward functions, such as scores from human preference models. We first show that it is possible to backpropagate the reward function gradient through the full samp…
What is the best paradigm to recognize objects---discriminative inference (fast but potentially prone to shortcut learning) or using a generative model (slow but potentially more robust)? We build on recent advances in generative modeling that turn text-to-image models into classifiers. This allows…
The excellent generative capabilities of text-to-image diffusion models suggest they learn informative representations of image-text data. However, what knowledge their representations capture is not fully understood, and they have not been thoroughly explored on downstream tasks. We investigate dif…
One of the major challenges that autonomous vehicles (AVs) face in an urban setting is communicating with other road users such as pedestrians. In this work, we investigated with what expressive behaviors we can endow AVs such that pedestrians readily recognize the underlying intent of the vehicles’
Dynamic evaluation of language models (LMs) adapts model parameters at test time using gradient information from previous tokens and substantially improves LM performance. However, it requires over 3x more compute than standard inference. We present Fast Weight Layers (FWLs), a neural component that…
Masked language modeling (MLM) pre-training methods such as BERT corrupt the input by replacing some tokens with [MASK] and then train a model to reconstruct the original tokens. While they produce good results when transferred to downstream NLP tasks, they generally require large amounts of compute…
We present Cross-View Training (CVT), a simple but effective method for deep semi-supervised learning. On labeled examples, the model is trained with standard cross-entropy loss. On an unlabeled example, the model first performs inference (acting as a "teacher") to produce soft targets. The model th…