Obtaining labeled data to train a model for a task of interest is often expensive. Prior work shows training models on multitask data augmented with task descriptions (prompts) effectively transfers knowledge to new tasks. Towards efficiently building task-specific models, we assume access to a small number (32-1000) of unlabeled target-task examples and use those to retrieve the most similar labeled examples from a large pool of multitask data augmented with prompts. Compared to the current practice of finetuning models on uniformly sampled prompted multitask data (e.g.: FLAN, T0), our approach of finetuning on cross-task nearest neighbors is significantly more data-efficient. Using only 2% of the data from the P3 pool without any labeled target-task data, our models outperform strong baselines trained on all available data by 3-30% on 12 out of 14 datasets representing held-out tasks including legal and scientific document QA. Similarly, models trained on cross-task nearest neighbors from SuperNaturalInstructions, representing about 5% of the pool, obtain comparable performance to state-of-the-art models on 12 held-out tasks from that pool. Moreover, the models produced by our approach also provide a better initialization than single multitask finetuned models for few-shot finetuning on target-task data, as shown by a 2-23% relative improvement over few-shot finetuned T0-3B models on 8 datasets.

This is part of the data recipe for Facebook's new large language model, which the company claims is able to detect and rein in harmful content in over 100 languages. Bumble uses similar technology to detect rude and unwanted messages in at least 15 languages. Google uses it for everything from translation to filtering newspaper comment sections. All have comparable recipes and the same dominant ingredient: English-language data. For years, social media companies have focused their automatic content detection and removal efforts more on content in English than the world's 7,000 other languages.

Microsoft has been in the video game business for over 20 years, but its Xbox console has been the perpetual runner-up. What's more, its huge $70bn merger with Activision Blizzard, a deal that would go a long way towards solving its content problem by buying it Call of Duty and much more, is currently being held up by UK regulators. Looking rather downtrodden on a recent episode of gaming show Xcast, Xbox boss Phil Spencer said that he was "not in the business of out-consoling Sony or out-consoling Nintendo", intimating that the long-running console war between Xbox, Nintendo and PlayStation was a losing battle for Microsoft. Instead of touting Xbox sales figures, for the past few years the company has presented its subscription service Game Pass as its metric for success: how many games it has, how many subscribers, how many hours they have played. It is betting on a post-console future, on being able to play games on any screen with the help of cloud servers – a streaming and subscription-centric future that has already transformed music, film and TV.

CANDECOMP/PARAFAC (CP) decomposition is the mostly used model to formulate the received tensor signal in a multi-domain massive multiple-input multiple-output (MIMO) system, as the receiver generally sums the components from different paths or users. To achieve accurate and low-latency channel estimation, good and fast CP decomposition algorithms are desired. The CP alternating least squares (CPALS) is the workhorse algorithm for calculating the CP decomposition. However, its performance depends on the initializations, and good starting values can lead to more efficient solutions. Existing initialization strategies are decoupled from the CPALS and are not necessarily favorable for solving the CP decomposition. To enhance the algorithm's speed and accuracy, this paper proposes a deep-learning-aided CPALS (DL-CPALS) method that uses a deep neural network (DNN) to generate favorable initializations. The proposed DL-CPALS integrates the DNN and CPALS to a model-based deep learning paradigm, where it trains the DNN to generate an initialization that facilitates fast and accurate CP decomposition. Moreover, benefiting from the CP low-rankness, the proposed method is trained using noisy data and does not require paired clean data. The proposed DL-CPALS is applied to millimeter wave MIMO orthogonal frequency division multiplexing (mmWave MIMO-OFDM) channel estimation. Experimental results demonstrate the significant improvements of the proposed method in terms of both speed and accuracy for CP decomposition and channel estimation.

Generalization to unseen tasks is an important ability for few-shot learners to achieve better zero-/few-shot performance on diverse tasks. However, such generalization to vision-language tasks including grounding and generation tasks has been under-explored; existing few-shot VL models struggle to handle tasks that involve object grounding and multiple images such as visual commonsense reasoning or NLVR2. In this paper, we introduce GRILL, GRounded vIsion Language aLigning, a novel VL model that can be generalized to diverse tasks including visual question answering, captioning, and grounding tasks with no or very few training instances. Specifically, GRILL learns object grounding and localization by exploiting object-text alignments, which enables it to transfer to grounding tasks in a zero-/few-shot fashion. We evaluate our model on various zero-/few-shot VL tasks and show that it consistently surpasses the state-of-the-art few-shot methods.

Representing texts as information about entities has long been deemed effective in event reasoning. We propose OpenPI2.0, an improved dataset for tracking entity states in procedural texts. OpenPI2.0 features not only canonicalized entities that facilitate evaluation, but also salience annotations including both manual labels and automatic predictions. Regarding entity salience, we provide a survey on annotation subjectivity, modeling feasibility, and downstream applications in tasks such as question answering and classical planning.

Visual language such as charts and plots is ubiquitous in the human world. Comprehending plots and charts requires strong reasoning skills. Prior state-of-the-art (SOTA) models require at least tens of thousands of training examples and their reasoning capabilities are still much limited, especially on complex human-written queries. This paper presents the first one-shot solution to visual language reasoning. We decompose the challenge of visual language reasoning into two steps: (1) plot-to-text translation, and (2) reasoning over the translated text. The key in this method is a modality conversion module, named as DePlot, which translates the image of a plot or chart to a linearized table. The output of DePlot can then be directly used to prompt a pretrained large language model (LLM), exploiting the few-shot reasoning capabilities of LLMs. To obtain DePlot, we standardize the plot-to-table task by establishing unified task formats and metrics, and train DePlot end-to-end on this task. DePlot can then be used off-the-shelf together with LLMs in a plug-and-play fashion. Compared with a SOTA model finetuned on more than >28k data points, DePlot+LLM with just one-shot prompting achieves a 24.0% improvement over finetuned SOTA on human-written queries from the task of chart QA.

To study social, economic, and historical questions, researchers in the social sciences and humanities have started to use increasingly large unstructured textual datasets. While recent advances in NLP provide many tools to efficiently process such data, most existing approaches rely on generic solutions whose performance and suitability for domain-specific tasks is not well understood. This work presents an attempt to bridge this domain gap by exploring the use of modern Entity Linking approaches for the enrichment of museum collection data. We collect a dataset comprising of more than 1700 texts annotated with 7,510 mention-entity pairs, evaluate some off-the-shelf solutions in detail using this dataset and finally fine-tune a recent end-to-end EL model on this data. We show that our fine-tuned model significantly outperforms other approaches currently available in this domain and present a proof-of-concept use case of this model. We release our dataset and our best model.

Deep learning models developed for time-series associated tasks have become more widely researched nowadays. However, due to the unintuitive nature of time-series data, the interpretability problem -- where we understand what is under the hood of these models -- becomes crucial. The advancement of similar studies in computer vision has given rise to many post-hoc methods, which can also shed light on how to explain time-series models. In this paper, we present a wide range of post-hoc interpretation methods for time-series models based on backpropagation, perturbation, and approximation. We also want to bring focus onto inherently interpretable models, a novel category of interpretation where human-understandable information is designed within the models. Furthermore, we introduce some common evaluation metrics used for the explanations, and propose several directions of future researches on the time-series interpretability problem. As a highlight, our work summarizes not only the well-established interpretation methods, but also a handful of fairly recent and under-developed techniques, which we hope to capture their essence and spark future endeavours to innovate and improvise.

Mathematical symbol definition extraction is important for improving scholarly reading interfaces and scholarly information extraction (IE). However, the task poses several challenges: math symbols are difficult to process as they are not composed of natural language morphemes; and scholarly papers often contain sentences that require resolving complex coordinate structures. We present SymDef, an English language dataset of 5,927 sentences from full-text scientific papers where each sentence is annotated with all mathematical symbols linked with their corresponding definitions. This dataset focuses specifically on complex coordination structures such as "respectively" constructions, which often contain overlapping definition spans. We also introduce a new definition extraction method that masks mathematical symbols, creates a copy of each sentence for each symbol, specifies a target symbol, and predicts its corresponding definition spans using slot filling. Our experiments show that our definition extraction model significantly outperforms RoBERTa and other strong IE baseline systems by 10.9 points with a macro F1 score of 84.82. With our dataset and model, we can detect complex definitions in scholarly documents to make scientific writing more readable.