A core process in human cognition is analogical mapping: the ability to identify a similar relational structure between different situations. We introduce a novel task, Visual Analogies of Situation Recognition, adapting the classical word-analogy task into the visual domain. Given a triplet of images, the task is to select an image candidate B' that completes the analogy (A to A' is like B to what?). Unlike previous work on visual analogy that focused on simple image transformations, we tackle complex analogies requiring understanding of scenes. We leverage situation recognition annotations and the CLIP model to generate a large set of 500k candidate analogies. Crowdsourced annotations for a sample of the data indicate that humans agree with the dataset label ~80% of the time (chance level 25%). Furthermore, we use human annotations to create a gold-standard dataset of 3,820 validated analogies. Our experiments demonstrate that state-of-the-art models do well when distractors are chosen randomly (~86%), but struggle with carefully chosen distractors (~53%, compared to 90% human accuracy). We hope our dataset will encourage the development of new analogy-making models. Website: https://vasr-dataset.github.io/

虽然视觉和语言模型在视觉问题回答等任务上表现良好，但在基本的人类常识性推理技能方面，它们会挣扎。在这项工作中，我们介绍了Winogavil：在线游戏，以收集视觉和语言协会（例如，狼人到满月），用作评估最先进模型的动态基准。受欢迎的纸牌游戏代号的启发，Spymaster提供了与几个视觉候选者相关的文本提示，另一个玩家必须识别它们。人类玩家因创建对竞争对手AI模型而具有挑战性的联想而获得了回报，但仍然可以由其他人类玩家解决。我们使用游戏来收集3.5k实例，发现它们对人类的直观（> 90％的Jaccard索引），但对最先进的AI模型充满挑战，其中最佳模型（Vilt）的得分为52％ ，成功的位置在视觉上是显着的。我们的分析以及我们从玩家那里收集的反馈表明，收集的关联需要多种推理技能，包括一般知识，常识，抽象等。我们发布数据集，代码和交互式游戏，旨在允许未来的数据收集，可用于开发具有更好关联能力的模型。

For applications that require processing large amounts of text at inference time, Large Language Models (LLMs) are handicapped by their limited context windows, which are typically 2048 tokens. In-context learning, an emergent phenomenon in LLMs in sizes above a certain parameter threshold, constitutes one significant example because it can only leverage training examples that fit into the context window. Existing efforts to address the context window limitation involve training specialized architectures, which tend to be smaller than the sizes in which in-context learning manifests due to the memory footprint of processing long texts. We present Parallel Context Windows (PCW), a method that alleviates the context window restriction for any off-the-shelf LLM without further training. The key to the approach is to carve a long context into chunks (``windows'') that fit within the architecture, restrict the attention mechanism to apply only within each window, and re-use the positional embeddings among the windows. We test the PCW approach on in-context learning with models that range in size between 750 million and 178 billion parameters, and show substantial improvements for tasks with diverse input and output spaces. Our results motivate further investigation of Parallel Context Windows as a method for applying off-the-shelf LLMs in other settings that require long text sequences.

Models trained from real-world data tend to imitate and amplify social biases. Although there are many methods suggested to mitigate biases, they require a preliminary information on the types of biases that should be mitigated (e.g., gender or racial bias) and the social groups associated with each data sample. In this work, we propose a debiasing method that operates without any prior knowledge of the demographics in the dataset, detecting biased examples based on an auxiliary model that predicts the main model's success and down-weights them during the training process. Results on racial and gender bias demonstrate that it is possible to mitigate social biases without having to use a costly demographic annotation process.

Dual encoders are now the dominant architecture for dense retrieval. Yet, we have little understanding of how they represent text, and why this leads to good performance. In this work, we shed light on this question via distributions over the vocabulary. We propose to interpret the vector representations produced by dual encoders by projecting them into the model's vocabulary space. We show that the resulting distributions over vocabulary tokens are intuitive and contain rich semantic information. We find that this view can explain some of the failure cases of dense retrievers. For example, the inability of models to handle tail entities can be explained via a tendency of the token distributions to forget some of the tokens of those entities. We leverage this insight and propose a simple way to enrich query and passage representations with lexical information at inference time, and show that this significantly improves performance compared to the original model in out-of-domain settings.

A household robot should be able to navigate to target locations without requiring users to first annotate everything in their home. Current approaches to this object navigation challenge do not test on real robots and rely on expensive semantically labeled 3D meshes. In this work, our aim is an agent that builds self-supervised models of the world via exploration, the same as a child might. We propose an end-to-end self-supervised embodied agent that leverages exploration to train a semantic segmentation model of 3D objects, and uses those representations to learn an object navigation policy purely from self-labeled 3D meshes. The key insight is that embodied agents can leverage location consistency as a supervision signal - collecting images from different views/angles and applying contrastive learning to fine-tune a semantic segmentation model. In our experiments, we observe that our framework performs better than other self-supervised baselines and competitively with supervised baselines, in both simulation and when deployed in real houses.

This paper investigates models of event implications. Specifically, how well models predict entity state-changes, by targeting their understanding of physical attributes. Nominally, Large Language models (LLM) have been exposed to procedural knowledge about how objects interact, yet our benchmarking shows they fail to reason about the world. Conversely, we also demonstrate that existing approaches often misrepresent the surprising abilities of LLMs via improper task encodings and that proper model prompting can dramatically improve performance of reported baseline results across multiple tasks. In particular, our results indicate that our prompting technique is especially useful for unseen attributes (out-of-domain) or when only limited data is available.

Large language models (LLMs) have been shown to be able to perform new tasks based on a few demonstrations or natural language instructions. While these capabilities have led to widespread adoption, most LLMs are developed by resource-rich organizations and are frequently kept from the public. As a step towards democratizing this powerful technology, we present BLOOM, a 176B-parameter open-access language model designed and built thanks to a collaboration of hundreds of researchers. BLOOM is a decoder-only Transformer language model that was trained on the ROOTS corpus, a dataset comprising hundreds of sources in 46 natural and 13 programming languages (59 in total). We find that BLOOM achieves competitive performance on a wide variety of benchmarks, with stronger results after undergoing multitask prompted finetuning. To facilitate future research and applications using LLMs, we publicly release our models and code under the Responsible AI License.

In dense neighborhoods, there are often dozens of homes in close proximity. This can either be a tight city-block with many single-family homes (SFHs), or a multiple dwelling units (MDU) complex (such as a big apartment building or condominium). Each home in such a neighborhood (either a SFH or a single unit in a MDU complex) has its own Wi-Fi access point (AP). Because there are few (typically 2 or 3) non-overlapping radio channels for Wi-Fi, neighboring homes may find themselves sharing a channel and competing over airtime, which may cause bad experience of slow internet (long latency, buffering while streaming movies, etc.). Wi-Fi optimization over all the APs in a dense neighborhood is highly desired to provide the best user experience. We present a method for Wi-Fi channel selection in a centralized way for all the APs in a dense neighborhood. We describe how to use recent observations to estimate the potential-pain matrix - for each pair of APs, how much Wi-Fi-pain would they cause each other if they were on the same channel. We formulate an optimization problem - finding a channel allocation (which channel each home should use) that minimizes the total Wi-Fi-pain in the neighborhood. We design an optimization algorithm that uses gradient descent over a neural network to solve the optimization problem. We describe initial results from offline experiments comparing our optimization solver to an off-the-shelf mixed-integer-programming solver. In our experiments we show that the off-the-shelf solver manages to find a better (lower total pain) solution on the train data (from the recent days), but our neural-network solver generalizes better - it finds a solution that achieves lower total pain for the test data (tomorrow).

The field of emergent communication aims to understand the characteristics of communication as it emerges from artificial agents solving tasks that require information exchange. Communication with discrete messages is considered a desired characteristic, for both scientific and applied reasons. However, training a multi-agent system with discrete communication is not straightforward, requiring either reinforcement learning algorithms or relaxing the discreteness requirement via a continuous approximation such as the Gumbel-softmax. Both these solutions result in poor performance compared to fully continuous communication. In this work, we propose an alternative approach to achieve discrete communication -- quantization of communicated messages. Using message quantization allows us to train the model end-to-end, achieving superior performance in multiple setups. Moreover, quantization is a natural framework that runs the gamut from continuous to discrete communication. Thus, it sets the ground for a broader view of multi-agent communication in the deep learning era.