Recent research has shown that large language models pretrained using unsupervised approaches can achieve significant performance improvement on many downstream tasks. Typically when adapting these language models to downstream tasks, like a classification or regression task, we employ a fine-tuning paradigm in which the sentence representation from the language model is input to a task-specific head; the model is then fine-tuned end-to-end. However, with the emergence of models like GPT-3, prompt-based fine-tuning has been proven to be a successful approach for few-shot tasks. Inspired by this work, we study discrete prompt technologies in practice. There are two issues that arise with the standard prompt approach. First, it can overfit on the prompt template. Second, it requires manual effort to formulate the downstream task as a language model problem. In this paper, we propose an improvement to prompt-based fine-tuning that addresses these two issues. We refer to our approach as DynaMaR – Dynamic Prompt with Mask Token Representation. Results show that DynaMaR can achieve an average improvement of 10% in few-shot settings and improvement of 3.7% in data-rich settings over the standard fine-tuning approach on four e-commerce applications.

Multi-task learning (MTL) aims to solve multiple tasks jointly by sharing a base representation among them. This can lead to more efficient learning and better generalization, as compared to learning each task individually. However, one issue that often arises in MTL is the convergence speed between tasks varies due to differences in task difficulty, so it can be a challenge to simultaneously achieve the best performance on all tasks with a single model checkpoint. Various techniques have been proposed to address discrepancies in task convergence rate, including weighting the per-task losses and modifying task gradients. In this work, we propose a novel approach that avoids the problem of requiring all tasks to converge at the same rate, but rather allows for “asynchronous” convergence among the tasks where each task can converge on its own schedule. As our main contribution, we monitor per-task validation metrics and switch to a knowledge distillation loss once a task has converged instead of continuing to train on the true labels. This prevents the model from overfitting on converged tasks while it learns the remaining tasks. We evaluate the proposed method in two 5-task MTL setups consisting of internal e-commerce datasets. The results show that our method consistently outperforms existing loss weighting and gradient balancing approaches, achieving average improvements of 0.9% and 1.5% over the best performing baseline model in the two setups, respectively.

While systems using the Neural Network-based Machine Translation (NMT) paradigm achieve the highest scores on recent shared tasks, phrase-based (PBMT) systems, rule-based (RBMT) systems and other systems may get better results for individual examples. Therefore, combined systems should achieve the best results for MT, particularly if the system combination method can take advantage of the strengths of each paradigm. In this paper, we describe a system that predicts whether a NMT, PBMT or RBMT will get the best Spanish translation result for a particular English sentence in DGT-TM 20161. Then we use fuzzy-match repair (FMR) as a mechanism to show that the combined system outperforms individual systems in a black-box machine translation setting.

Event detection (ED) and word sense disambiguation (WSD) are two similar tasks in that they both involve identifying the classes (i.e. event types or word senses) of some word in a given sentence. It is thus possible to extract the knowledge hidden in the data for WSD, and utilize it to improve the performance on ED. In this work, we propose a method to transfer the knowledge learned on WSD to ED by matching the neural representations learned for the two tasks. Our experiments on two widely used datasets for ED demonstrate the effectiveness of the proposed method.

The EmoInt-2017 task aims to determine a continuous numerical value representing the intensity to which an emotion is expressed in a tweet. Compared to classification tasks that identify 1 among n emotions for a tweet, the present task can provide more fine-grained (real-valued) sentiment analysis. This paper presents a system that uses a bi-directional LSTM-CNN model to complete the competition task. Combining bi-directional LSTM and CNN, the prediction process considers both global information in a tweet and local important information. The proposed method ranked sixth among twenty-one teams in terms of Pearson Correlation Coefficient.