We propose a benchmark to assess the capability of large language models to reason with conventional metaphors. Our benchmark combines the previously isolated topics of metaphor detection and commonsense reasoning into a single task that requires a model to make inferences by accurately selecting between the literal and metaphorical register. We examine the performance of state-of-the-art pre-trained models on binary-choice tasks and find a large discrepancy between the performance of small and very large models, going from chance to near-human level. We also analyse the largest model in a generative setting and find that although human performance is approached, careful multiple-shot prompting is required.

Figurative language understanding has been recently framed as a recognizing textual entailment (RTE) task (a.k.a. natural language inference, or NLI). However, similar to classical RTE/NLI datasets, the current benchmarks suffer from spurious correlations and annotation artifacts. To tackle this problem, work on NLI has built explanation-based datasets such as e-SNLI, allowing us to probe whether language models are right for the right reasons.Yet no such data exists for figurative language, making it harder to assess genuine understanding of such expressions. To address this issue, we release FLUTE, a dataset of 9,000 figurative NLI instances with explanations, spanning four categories: Sarcasm, Simile, Metaphor, and Idioms. We collect the data through a model-in-the-loop framework based on GPT-3, crowd workers, and expert annotators. We show how utilizing GPT-3 in conjunction with human annotators (novices and experts) can aid in scaling up the creation of datasets even for such complex linguistic phenomena as figurative language. The baseline performance of the T5 model fine-tuned on FLUTE shows that our dataset can bring us a step closer to developing models that understand figurative language through textual explanations.

A bottleneck to developing Semantic Parsing (SP) models is the need for a large volume of human-labeled training data. Given the complexity and cost of human annotation for SP, labeled data is often scarce, particularly in multilingual settings. Large Language Models (LLMs) excel at SP given only a few examples, however LLMs are unsuitable for runtime systems which require low latency. In this work, we propose CLASP, a simple method to improve low-resource SP for moderate-sized models: we generate synthetic data from AlexaTM 20B to augment the training set for a model 40x smaller (500M parameters). We evaluate on two datasets in low-resource settings: English PIZZA, containing either 348 or 16 real examples, and mTOP cross-lingual zero-shot, where training data is available only in English, and the model must generalize to four new languages. On both datasets, we show significant improvements over strong baseline methods.

Many NLP classification tasks, such as sexism/racism detection or toxicity detection, are based on human values. Yet, human values can vary under diverse cultural conditions. Therefore, we introduce a framework for value-aligned classification that performs prediction based on explicitly written human values in the command. Along with the task, we propose a practical approach that distills value-aligned knowledge from large-scale language models (LLMs) to construct value-aligned classifiers in two steps. First, we generate value-aligned training data from LLMs by prompt-based few-shot learning. Next, we fine-tune smaller classification models with the generated data for the task. Empirical results show that our VA-Models surpass multiple baselines by at least 15.56% on the F1-score, including few-shot learning with OPT-175B and existing text augmentation methods. We suggest that using classifiers with explicit human value input improves both inclusivity & explainability in AI.

The recently proposed capability-based NLP testing allows model developers to test the functional capabilities of NLP models, revealing functional failures that cannot be detected by the traditional heldout mechanism. However, existing work on capability-based testing requires extensive manual efforts and domain expertise in creating the test cases. In this paper, we investigate a low-cost approach for the test case generation by leveraging the GPT-3 engine. We further propose to use a classifier to remove the invalid outputs from GPT-3 and expand the outputs into templates to generate more test cases. Our experiments show that TestAug has three advantages over the existing work on behavioral testing: (1) TestAug can find more bugs than existing work; (2) The test cases in TestAug are more diverse; and (3) TestAug largely saves the manual efforts in creating the test suites. The code and data for TestAug can be found at our project website (https://guanqun-yang.github.io/testaug/) and GitHub (https://github.com/guanqun-yang/testaug).

Most downstream adaptation methods tune all or part of the parameters of pre-trained models (PTMs) through gradient descent, where the tuning cost increases linearly with the growth of the model size. By contrast, gradient-free methods only require the forward computation of the PTM to tune the prompt, retaining the benefits of efficient tuning and deployment. Though, past work on gradient-free tuning often introduces gradient descent to seek a good initialization of prompt and lacks versatility across tasks and PTMs. In this paper, we present BBTv2, an improved version of Black-Box Tuning, to drive PTMs for few-shot learning. We prepend continuous prompts to every layer of the PTM and propose a divide-and-conquer gradient-free algorithm to optimize the prompts at different layers alternately. Extensive experiments across various tasks and PTMs show that BBTv2 can achieve comparable performance to full model tuning and state-of-the-art parameter-efficient methods (e.g., Adapter, LoRA, BitFit, etc.) under few-shot settings while maintaining much fewer tunable parameters.

At DeepMind, we value diversity of experience, knowledge, backgrounds and perspectives and harness these qualities to create extraordinary impact. We are committed to equal employment opportunity regardless of sex, race, religion or belief, ethnic or national origin, disability, age, citizenship, marital, domestic or civil partnership status, sexual orientation, gender identity, pregnancy, parental or related condition (including breastfeeding) or any other basis as protected by applicable law. If you have a disability or additional need that requires accommodation, please do not hesitate to let us know. We are looking for a Technical Program Manager to support a newly formed team looking after data at DeepMind! This is an opportunity to focus on our efforts with dataset acquisition.

A pair of researchers have found a more efficient way to multiply grids of numbers, beating a record set just a week ago by the artificial intelligence firm DeepMind. The company revealed on 5 October that its AI software had beaten a record that had stood for more than 50 years for the matrix multiplication problem – a common operation in all sorts of software where grids of numbers are multiplied by each other. DeepMind's paper revealed a new method for multiplying two …

To further strengthen our commitment to providing industry-leading coverage of data technology, VentureBeat is excited to welcome Andrew Brust and Tony Baer as regular contributors. Is there a better way to build a search tool that produces more highly relevant results than just using keyword-based techniques? That's one of the many questions that former Google staffers Amr Awadallah (CEO), Amin Ahmad (CTO) and Tallat Shafaat (chief architect) wanted to solve with their new startup, which has been in stealth under the name ZIR AI. Today, ZIR AI is emerging from stealth under the name Vectara, with the help of $20 million in seed funding, and availability of the company's neural search-as-a-service technology. The foundational premise of Vectara is that artificial intelligence (AI)-based large language models (LLMs) combined with natural language processing (NLP), data integration pipelines and vector techniques can create a neural network that is useful for multiple use cases, including search.

Learning visual representations from natural language supervision has recently shown great promise in a number of pioneering works. In general, these language-augmented visual models demonstrate strong transferability to a variety of datasets and tasks. However, it remains challenging to evaluate the transferablity of these models due to the lack of easy-to-use evaluation toolkits and public benchmarks. To tackle this, we build ELEVATER (Evaluation of Language-augmented Visual Task-level Transfer), the first benchmark and toolkit for evaluating(pre-trained) language-augmented visual models. ELEVATER is composed of three components. (i) Datasets. As downstream evaluation suites, it consists of 20 image classification datasets and 35 object detection datasets, each of which is augmented with external knowledge. (ii) Toolkit. An automatic hyper-parameter tuning toolkit is developed to facilitate model evaluation on downstream tasks. (iii) Metrics. A variety of evaluation metrics are used to measure sample-efficiency (zero-shot and few-shot) and parameter-efficiency (linear probing and full model fine-tuning). ELEVATER is a platform for Computer Vision in the Wild (CVinW), and is publicly released at at https://computer-vision-in-the-wild.github.io/ELEVATER/