Pre-trained Transformer models such as BERT have shown great success in a wide range of applications, but at the cost of substantial increases in model complexity. Quantization-aware training (QAT) is a promising method to lower the implementation cost and energy consumption. However, aggressive quantization below 2-bit causes considerable accuracy degradation due to unstable convergence, especially when the downstream dataset is not abundant. This work proposes a proactive knowledge distillation method called Teacher Intervention (TI) for fast converging QAT of ultra-low precision pre-trained Transformers. TI intervenes layer-wise signal propagation with the intact signal from the teacher to remove the interference of propagated quantization errors, smoothing loss surface of QAT and expediting the convergence. Furthermore, we propose a gradual intervention mechanism to stabilize the recovery of subsections of Transformer layers from quantization. The proposed schemes enable fast convergence of QAT and improve the model accuracy regardless of the diverse characteristics of downstream fine-tuning tasks. We demonstrate that TI consistently achieves superior accuracy with significantly lower fine-tuning iterations on well-known Transformers of natural language processing as well as computer vision compared to the state-of-the-art QAT methods.

Text classification with extremely weak supervision (EWS) imposes stricter supervision constraints compared to regular weakly supervise classification. Absolutely no labeled training samples or hand-crafted rules specific to the evaluation data are allowed. Such restrictions limit state-of-the-art EWS classification methods to indirect weak labeling techniques that assign unnatural label uncertainty estimates. We present PLAT, a framework that creates weak labels by leveraging recent developments in zero-shot text classification. PLAT employs models trained for sub-tasks other than classification to label documents. Most importantly, PLAT refrains from assigning overly confident weak labels and improves soft-label training performance for downstream classifiers. Classifiers trained with PLAT significantly outperform those trained on weak labels generated by the previous state-of-the-art in extremely weakly supervised text classification.

In weakly-supervised text classification, only label names act as sources of supervision. Predominant approaches to weakly-supervised text classification utilize a two-phase framework, where test samples are first assigned pseudo-labels and are then used to train a neural text classifier. In most previous work, the pseudo-labeling step is dependent on obtaining seed words that best capture the relevance of each class label. We present LIME, a framework for weakly-supervised text classification that entirely replaces the brittle seed-word generation process with entailment-based pseudo-classification. We find that combining weakly-supervised classification and textual entailment mitigates shortcomings of both, resulting in a more streamlined and effective classification pipeline. With just an off-the-shelf textual entailment model, LIME outperforms recent baselines in weakly-supervised text classification and achieves state-of-the-art in 4 benchmarks.

To mitigate the lack of diverse dialogue summarization datasets in academia, we present methods to utilize non-dialogue summarization data for enhancing dialogue summarization systems. We apply transformations to document summarization data pairs to create training data that better befit dialogue summarization. The suggested transformations also retain desirable properties of non-dialogue datasets, such as improved faithfulness to the source text. We conduct extensive experiments across both English and Korean to verify our approach. Although absolute gains in ROUGE naturally plateau as more dialogue summarization samples are introduced, utilizing non-dialogue data for training significantly improves summarization performance in zero- and few-shot settings and enhances faithfulness across all training regimes.

We advance the state-of-the-art in unsupervised abstractive dialogue summarization by utilizing multi-sentence compression graphs. Starting from well-founded assumptions about word graphs, we present simple but reliable path-reranking and topic segmentation schemes. Robustness of our method is demonstrated on datasets across multiple domains, including meetings, interviews, movie scripts, and day-to-day conversations. We also identify possible avenues to augment our heuristic-based system with deep learning. We open-source our code, to provide a strong, reproducible baseline for future research into unsupervised dialogue summarization.

Text variational autoencoders (VAEs) are notorious for posterior collapse, a phenomenon where the model’s decoder learns to ignore signals from the encoder. Because posterior collapse is known to be exacerbated by expressive decoders, Transformers have seen limited adoption as components of text VAEs. Existing studies that incorporate Transformers into text VAEs (Li et al., 2020; Fang et al., 2021) mitigate posterior collapse using massive pretraining, a technique unavailable to most of the research community without extensive computing resources. We present a simple two-phase training scheme to convert a sequence-to-sequence Transformer into a VAE with just finetuning. The resulting language model is competitive with massively pretrained Transformer-based VAEs in some internal metrics while falling short on others. To facilitate training we comprehensively explore the impact of common posterior collapse alleviation techniques in the literature. We release our code for reproducability.