Po-Nien Kung


2023

pdf bib
Active Instruction Tuning: Improving Cross-Task Generalization by Training on Prompt Sensitive Tasks
Po-Nien Kung | Fan Yin | Di Wu | Kai-Wei Chang | Nanyun Peng
Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing

Instruction tuning (IT) achieves impressive zero-shot generalization results by training large language models (LLMs) on a massive amount of diverse tasks with instructions. However, how to select new tasks to improve the performance and generalizability of IT models remains an open question. Training on all existing tasks is impractical due to prohibiting computation requirements, and randomly selecting tasks can lead to suboptimal performance. In this work, we propose active instruction tuning based on prompt uncertainty, a novel framework to identify informative tasks, and then actively tune the models on the selected tasks. We represent the informativeness of new tasks with the disagreement of the current model outputs over perturbed prompts. Our experiments on NIV2 and Self-Instruct datasets demonstrate that our method consistently outperforms other baseline strategies for task selection, achieving better out-of-distribution generalization with fewer training tasks. Additionally, we introduce a task map that categorizes and diagnoses tasks based on prompt uncertainty and prediction probability. We discover that training on ambiguous (prompt-uncertain) tasks improves generalization while training on difficult (prompt-certain and low-probability) tasks offers no benefit, underscoring the importance of task selection for instruction tuning.

pdf bib
Hierarchical Representations in Dense Passage Retrieval for Question-Answering
Philipp Ennen | Federica Freddi | Chyi-Jiunn Lin | Po-Nien Kung | RenChu Wang | Chien-Yi Yang | Da-shan Shiu | Alberto Bernacchia
Proceedings of the Sixth Fact Extraction and VERification Workshop (FEVER)

An approach to improve question-answering performance is to retrieve accompanying information that contains factual evidence matching the question. These retrieved documents are then fed into a reader that generates an answer. A commonly applied retriever is dense passage retrieval. In this retriever, the output of a transformer neural network is used to query a knowledge database for matching documents. Inspired by the observation that different layers of a transformer network provide rich representations with different levels of abstraction, we hypothesize that useful queries can be generated not only at the output layer, but at every layer of a transformer network, and that the hidden representations of different layers may combine to improve the fetched documents for reader performance. Our novel approach integrates retrieval into each layer of a transformer network, exploiting the hierarchical representations of the input question. We show that our technique outperforms prior work on downstream tasks such as question answering, demonstrating the effectiveness of our approach.

pdf bib
Do Models Really Learn to Follow Instructions? An Empirical Study of Instruction Tuning
Po-Nien Kung | Nanyun Peng
Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)

Recent works on instruction tuning (IT) have achieved great performance with zero-shot generalizability to unseen tasks. With additional context (e.g., task definition, examples) provided to models for fine-tuning, they achieved much higher performance than untuned models. Despite impressive performance gains, what models learn from IT remains understudied. In this work, we analyze how models utilize instructions during IT by comparing model training with altered vs. original instructions. Specifically, we create simplified task definitions by removing all semantic components and only leaving the output space information, and delusive examples that contain incorrect input-output mapping. Our experiments show that models trained on simplified task definition or delusive examples can achieve comparable performance to the ones trained on the original instructions and examples. Furthermore, we introduce a random baseline to perform zeroshot classification tasks, and find it achieves similar performance (42.6% exact-match) as IT does (43% exact-match) in low resource setting, while both methods outperform naive T5 significantly (30% per exact-match). Our analysis provides evidence that the impressive performance gain of current IT models can come from picking up superficial patterns, such as learning the output format and guessing. Our study highlights the urgent need for more reliable IT methods and evaluation.

2021

pdf bib
Efficient Multi-Task Auxiliary Learning: Selecting Auxiliary Data by Feature Similarity
Po-Nien Kung | Sheng-Siang Yin | Yi-Cheng Chen | Tse-Hsuan Yang | Yun-Nung Chen
Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing

Multi-task auxiliary learning utilizes a set of relevant auxiliary tasks to improve the performance of a primary task. A common usage is to manually select multiple auxiliary tasks for multi-task learning on all data, which raises two issues: (1) selecting beneficial auxiliary tasks for a primary task is nontrivial; (2) when the auxiliary datasets are large, training on all data becomes time-expensive and impractical. Therefore, this paper focuses on addressing these problems and proposes a time-efficient sampling method to select the data that is most relevant to the primary task. The proposed method allows us to only train on the most beneficial sub-datasets from the auxiliary tasks, achieving efficient multi-task auxiliary learning. The experiments on three benchmark datasets (RTE, MRPC, STS-B) show that our method significantly outperforms random sampling and ST-DNN. Also, by applying our method, the model can surpass fully-trained MT-DNN on RTE, MRPC, STS-B, using only 50%, 66%, and 1% of data, respectively.

2020

pdf bib
Zero-Shot Rationalization by Multi-Task Transfer Learning from Question Answering
Po-Nien Kung | Tse-Hsuan Yang | Yi-Cheng Chen | Sheng-Siang Yin | Yun-Nung Chen
Findings of the Association for Computational Linguistics: EMNLP 2020

Extracting rationales can help human understand which information the model utilizes and how it makes the prediction towards better interpretability. However, annotating rationales requires much effort and only few datasets contain such labeled rationales, making supervised learning for rationalization difficult. In this paper, we propose a novel approach that leverages the benefits of both multi-task learning and transfer learning for generating rationales through question answering in a zero-shot fashion. For two benchmark rationalization datasets, the proposed method achieves comparable or even better performance of rationalization without any supervised signal, demonstrating the great potential of zero-shot rationalization for better interpretability.