The Embodied AI community has made significant strides in visual navigation tasks, exploring targets from 3D coordinates, objects, language descriptions, and images. However, these navigation models often handle only a single input modality as the target. With the progress achieved so for, it is time to move towards universal navigation models capable of handling various goal types, enabling more effective user interaction with robots. To facilitate this goal, we propose GOAT-Bench, a benchmark for the universal navigation task referred to as GO to AnyThing (GOAT). In this task, the agent is directed to navigate to a sequence of targets specified by the category name, language description, or image in an open-vocabulary fashion. We benchmark monolithic RL and modular methods on the GOAT task, analyzing their performance across modalities, the role of explicit and implicit scene memories, their robustness to noise in goal specifications, and the impact of memory in lifelong scenarios.

We present a modern formulation of Embodied Question Answering (EQA) as the task of understanding an environment well enough to answer questions about it in natural language. An agent can achieve such an understanding by either drawing upon episodic memory, exemplified by agents on smart glasses, or by actively exploring the environment, as in the case of mobile robots. We accompany our formulation with OpenEQA - the first open-vocabulary benchmark dataset for EQA supporting both episodic memory and active exploration use cases. OpenEQA contains over 1600 high-quality human generated questions drawn from over 180 real-world environments. In addition to the dataset, we also provide an automatic LLM-powered evaluation protocol that has excellent correlation with human judgement. Using this dataset and evaluation protocol, we evaluate several state-of-the-art foundation models including GPT-4V, and find that they significantly lag behind human-level performance. Consequently, OpenEQA stands out as a straightforward, measurable, and practically rele-vant benchmark that poses a considerable challenge to current generation offoundation models. We hope this inspires and stimulates future research at the intersection of Embod-ied AI, conversational agents, and world models.

Computer vision datasets frequently contain spurious correlations between task-relevant labels and (easy to learn) latent task-irrelevant attributes (e.g. context). Models trained on such datasets learn "shortcuts" and underperform on bias-conflicting slices of data where the correlation does not hold. In this work, we study the problem of identifying such slices to inform downstream bias mitigation strategies. We propose First Amplify Correlations and Then Slice (FACTS), wherein we first amplify correlations to fit a simple bias-aligned hypothesis via strongly regularized empirical risk minimization. Next, we perform correlation-aware slicing via mixture modeling in bias-aligned feature space to discover underperforming data slices that capture distinct correlations. Despite its simplicity, our method considerably improves over prior work (by as much as 35% precision@10) in correlation bias identification across a range of diverse evaluation settings. Code: https://github.com/yvsriram/FACTS.

We propose an automated algorithm to stress-test a trained visual model by generating language-guided counterfactual test images (LANCE). Our method leverages recent progress in large language modeling and text-based image editing to augment an IID test set with a suite of diverse, realistic, and challenging test images without altering model weights. We benchmark the performance of a diverse set of pre-trained models on our generated data and observe significant and consistent performance drops. We further analyze model sensitivity across different types of edits, and demonstrate its applicability at surfacing previously unknown class-level model biases in ImageNet. Code is available at https://github.com/virajprabhu/lance.

Visual domain adaptation (DA) seeks to transfer trained models to unseen, unlabeled domains across distribution shift, but approaches typically focus on adapting convolutional neural network architectures initialized with supervised ImageNet representations. In this work, we shift focus to adapting modern architectures for object recognition -- the increasingly popular Vision Transformer (ViT) -- initialized with modern pretraining based on self-supervised learning (SSL). Inspired by the design of recent SSL approaches based on learning from partial image inputs generated via masking or cropping -- either by learning to predict the missing pixels, or learning representational invariances to such augmentations -- we propose PACMAC, a two-stage adaptation algorithm for self-supervised ViTs. PACMAC first performs in-domain SSL on pooled source and target data to learn task-discriminative features, and then probes the model's predictive consistency across a set of partial target inputs generated via a novel attention-conditioned masking strategy, to identify reliable candidates for self-training. Our simple approach leads to consistent performance gains over competing methods that use ViTs and self-supervised initializations on standard object recognition benchmarks. Our code is available at https://github.com/virajprabhu/PACMAC.

In this work, we present FaaSter, an Accelerated Functions as a Service (AFaaS) offering that unifies the function-as-a-service model with GPU acceleration resources. FaaSter provides an acceleration function library as a service, which in turn is provisioned on heterogeneous GPUs. To provide seamless access to these accelerated functions and ensure that each function has the best possible response time, we utilize GPU kernel slicing to split and execute an accelerated function instance across multiple heterogeneous GPUs. The central challenge is to be able to quickly decide the number of slices to split each function into and then map the slices to the right GPUs. To this end, we present a scheduling heuristic that is able to significantly reduce the average turn-around time of functions when compared to a non-sliceable full GPU scheduling approach. Our evaluation results show that the FaaSter scheduler achieves 62 % mean and up to 80 % improvement in average turn-around time and always performs equal to or better than non-sliceable GPU scheduling.

Current approaches for semantic image inpainting rely on deep neural networks (DNNs) that learn under full supervision, i.e., using a training set comprising pairs of (i) corrupted images with holes and (ii) corresponding uncorrupted images. However, for several real-world applications, obtaining large sets of uncorrupted images is challenging or infeasible. Current methods also rely on adversarial training involving min-max optimization that is prone to instability during learning. We propose a novel self-supervised image-inpainting DNN framework that can learn in both completely unsupervised and semi-supervised modes. Moreover, our DNN learning formulation bypasses adversarial training and, thereby, lends itself to more stable training. Results on the publicly available CelebA dataset show that our method, even when learning unsupervisedly, outperforms the state of the art that learns with full supervision.