Despite substantial progress in abstractive text summarization to generate fluent and informative texts, the factual inconsistency in the generated summaries remains an important yet challenging problem to be solved. In this paper, we construct causal graphs for abstractive text summarization and identify the intrinsic causes of the factual inconsistency, i.e., the language bias and irrelevancy bias, and further propose a debiasing framework, named CoFactSum, to alleviate the causal effects of these biases by counterfactual estimation. Specifically, the proposed CoFactSum provides two counterfactual estimation strategies, i.e., Explicit Counterfactual Masking with an explicit dynamic masking strategy, and Implicit Counterfactual Training with an implicit discriminative cross-attention mechanism. Meanwhile, we design a Debiasing Degree Adjustment mechanism to dynamically adapt the debiasing degree at each decoding step. Extensive experiments on two widely-used summarization datasets demonstrate the effectiveness of CoFactSum in enhancing the factual consistency of generated summaries compared with several baselines.
An important problem of the sequence-to-sequence neural models widely used in abstractive summarization is exposure bias. To alleviate this problem, re-ranking systems have been applied in recent years. Despite some performance improvements, this approach remains underexplored. Previous works have mostly specified the rank through the ROUGE score and aligned candidate summaries, but there can be quite a large gap between the lexical overlap metric and semantic similarity. In this paper, we propose a novel training method in which a re-ranker balances the lexical and semantic quality. We further newly define false positives in ranking and present a strategy to reduce their influence. Experiments on the CNN/DailyMail and XSum datasets show that our method can estimate the meaning of summaries without seriously degrading the lexical aspect. More specifically, it achieves an 89.67 BERTScore on the CNN/DailyMail dataset, reaching new state-of-the-art performance. Our code is publicly available at https://github.com/jeewoo1025/BalSum.
End-to-end spoken language understanding (SLU) remains elusive even with current large pretrained language models on text and speech, especially in multilingual cases. Machine translation has been established as a powerful pretraining objective on text as it enables the model to capture high-level semantics of the input utterance and associations between different languages, which is desired for speech models that work on lower-level acoustic frames. Motivated particularly by the task of cross-lingual SLU, we demonstrate that the task of speech translation (ST) is a good means of pretraining speech models for end-to-end SLU on both intra- and cross-lingual scenarios. By introducing ST, our models reach higher performance over baselines on monolingual and multilingual intent classification as well as spoken question answering using SLURP, MINDS-14, and NMSQA benchmarks. To verify the effectiveness of our methods, we also create new benchmark datasets from both synthetic and real sources, for speech summarization and low-resource/zero-shot transfer from English to French or Spanish. We further show the value of preserving knowledge for the ST pretraining task for better downstream performance, possibly using Bayesian transfer regularizers.
Based on the remarkable achievements of pre-trained language models in abstractive summarization, the copying mechanism has proved helpful by improving the factuality, stability, and overall performance. This work proposes PROM, a new PhRase-level cOpying Mechanism that enhances attention on n-grams, which can be applied to zero-shot summarization with pre-training. PROM adds an indicator layer to explicitly pick up tokens in n-gram that can be copied from the source, and calculates an auxiliary loss for the copying prediction. Empirical studies show that PROM makes significant improvements in fine-tuning on benchmarks. In zero-shot setting, PROM is utilized in the self-supervised pre-training on raw corpora and provides new general baselines on a wide range of summarization datasets. Further analysis shows that PROM performs more reasonable copying and contributes to faithfulness.
Debatepedia is a publicly available dataset consisting of arguments and counter-arguments on controversial topics that has been widely used for the single-document query-focused abstractive summarization task in recent years. However, it has been recently found that this dataset is limited by noise and even most queries in this dataset do not have any relevance to the respective document. In this paper, we present a methodology for cleaning the Debatepedia dataset by leveraging the generative power of large language models to make it suitable for query-focused abstractive summarization. More specifically, we harness the language generation capabilities of ChatGPT to regenerate its queries. We evaluate the effectiveness of the proposed ChatGPT annotated version of the Debatepedia dataset using several benchmark summarization models and demonstrate that the newly annotated version of Debatepedia outperforms the original dataset in terms of both query relevance as well as summary generation quality. We will make this annotated and cleaned version of the dataset publicly available.
Multimodal abstractive summarization for videos (MAS) requires generating a concise textual summary to describe the highlights of a video according to multimodal resources, in our case, the video content and its transcript. Inspired by the success of the large-scale generative pre-trained language model (GPLM) in generating high-quality textual content (e.g., summary), recent MAS methods have proposed to adapt the GPLM to this task by equipping it with the visual information, which is often obtained through a general-purpose visual feature extractor. However, the generally extracted visual features may overlook some summary-worthy visual information, which impedes model performance. In this work, we propose a novel approach to learning the summary-worthy visual representation that facilitates abstractive summarization. Our method exploits the summary-worthy information from both the cross-modal transcript data and the knowledge that distills from the pseudo summary. Extensive experiments on three public multimodal datasets show that our method outperforms all competing baselines. Furthermore, with the advantages of summary-worthy visual information, our model can have a significant improvement on small datasets or even datasets with limited training data.
With the explosive growth of textual information, summarization systems have become increasingly important. This work aims to concisely indicate the current state of the art in abstractive text summarization. As part of this, we outline the current paradigm shifts towards pre-trained encoder-decoder models and large autoregressive language models. Additionally, we delve further into the challenges of evaluating summarization systems and the potential of instruction-tuned models for zero-shot summarization. Finally, we provide a brief overview of how summarization systems are currently being integrated into commercial applications.
Language models (LMs) can generate hallucinations and incoherent outputs, which highlights their weak context dependency. Cache-LMs, which augment LMs with a memory of recent history, can increase context dependency and have shown remarkable performance in diverse language generation tasks. However, we find that even with training, the performance gain stemming from the cache component of current cache-LMs is suboptimal due to the misalignment between the current hidden states and those stored in the memory. In this work, we present HistAlign, a new training approach to ensure good cache alignment such that the model receives useful signals from the history. We first prove our concept on a simple and synthetic task where the memory is essential for correct predictions, and we show that the cache component of HistAlign is better aligned and improves overall performance. Next, we evaluate HistAlign on diverse downstream language generation tasks, including prompt continuation, abstractive summarization, and data-to-text. We demonstrate that HistAlign improves text coherence and faithfulness in open-ended and conditional generation settings respectively. HistAlign is also generalizable across different model families, showcasing its strength in improving context dependency of LMs in diverse scenarios. Our code is publicly available at https://github.com/meetdavidwan/histalign
Tailoring outputs from large language models, like ChatGPT, to implicit user preferences remains a challenge despite their impressive generative capabilities. In this paper, we propose a tri-agent generation pipeline comprising a generator, an instructor, and an editor to enhance output personalization. The generator produces an initial output, the instructor automatically generates editing instructions based on user preferences, and the editor refines the output to align with those preferences. The inference-only large language model (ChatGPT) serves as both the generator and editor, with a smaller model acting as the instructor to guide output generation. We train the instructor using editor-steered reinforcement learning, leveraging feedback from a large-scale editor model to optimize instruction generation. Experimental results on two abstractive summarization datasets demonstrate the effectiveness of our approach in generating outputs that better meet user expectations. Code is available at \url{https://github.com/Wendy-Xiao/chatgpt_editing_summ}
With direct access to human-written reference as memory, retrieval-augmented generation has achieved much progress in a wide range of text generation tasks. Since better memory would typically prompt better generation~(we define this as primal problem). The traditional approach for memory retrieval involves selecting memory that exhibits the highest similarity to the input. However, this method is constrained by the quality of the fixed corpus from which memory is retrieved. In this paper, by exploring the duality of the primal problem: better generation also prompts better memory, we propose a novel framework, selfmem, which addresses this limitation by iteratively employing a retrieval-augmented generator to create an unbounded memory pool and using a memory selector to choose one output as memory for the subsequent generation round. This enables the model to leverage its own output, referred to as self-memory, for improved generation. We evaluate the effectiveness of selfmem on three distinct text generation tasks: neural machine translation, abstractive text summarization, and dialogue generation, under two generation paradigms: fine-tuned small model and few-shot LLM. Our approach achieves state-of-the-art results in four directions in JRC-Acquis, XSum (50.3 ROUGE-1), and BigPatent (62.9 ROUGE-1), demonstrating the potential of self-memory in enhancing retrieval-augmented generation models. Furthermore, we conduct thorough analyses of each component in the selfmem framework to identify bottlenecks and provide insights for future research.
Recent pre-trained language models (PLMs) achieve promising results in existing abstractive summarization datasets. However, existing summarization benchmarks overlap in time with the standard pre-training corpora and finetuning datasets. Hence, the strong performance of PLMs may rely on the parametric knowledge that is memorized during pre-training and fine-tuning. Moreover, the knowledge memorized by PLMs may quickly become outdated, which affects the generalization performance of PLMs on future data. In this work, we propose TempoSum, a novel benchmark that contains data samples from 2010 to 2022, to understand the temporal generalization ability of abstractive summarization models. Through extensive human evaluation, we show that parametric knowledge stored in summarization models significantly affects the faithfulness of the generated summaries on future data. Moreover, existing faithfulness enhancement methods cannot reliably improve the faithfulness of summarization models on future data. Finally, we discuss several recommendations to the research community on how to evaluate and improve the temporal generalization capability of text summarization models.
Growing literature has shown that NLP systems may encode social biases; however, the political bias of summarization models remains relatively unknown. In this work, we use an entity replacement method to investigate the portrayal of politicians in automatically generated summaries of news articles. We develop an entity-based computational framework to assess the sensitivities of several extractive and abstractive summarizers to the politicians Donald Trump and Joe Biden. We find consistent differences in these summaries upon entity replacement, such as reduced emphasis of Trump's presence in the context of the same article and a more individualistic representation of Trump with respect to the collective US government (i.e., administration). These summary dissimilarities are most prominent when the entity is heavily featured in the source article. Our characterization provides a foundation for future studies of bias in summarization and for normative discussions on the ideal qualities of automatic summaries.
Human experts write summaries using different techniques, including extracting a sentence from the document and rewriting it, or fusing various information from the document to abstract it. These techniques are flexible and thus difficult to be imitated by any single method. To address this issue, we propose an adaptive model, GEMINI, that integrates a rewriter and a generator to mimic the sentence rewriting and abstracting techniques, respectively. GEMINI adaptively chooses to rewrite a specific document sentence or generate a summary sentence from scratch. Experiments demonstrate that our adaptive approach outperforms the pure abstractive and rewriting baselines on three benchmark datasets, achieving the best results on WikiHow. Interestingly, empirical results show that the human summary styles of summary sentences are consistently predictable given their context. We release our code and model at \url{https://github.com/baoguangsheng/gemini}.
Cross-lingual science journalism generates popular science stories of scientific articles different from the source language for a non-expert audience. Hence, a cross-lingual popular summary must contain the salient content of the input document, and the content should be coherent, comprehensible, and in a local language for the targeted audience. We improve these aspects of cross-lingual summary generation by joint training of two high-level NLP tasks, simplification and cross-lingual summarization. The former task reduces linguistic complexity, and the latter focuses on cross-lingual abstractive summarization. We propose a novel multi-task architecture - SimCSum consisting of one shared encoder and two parallel decoders jointly learning simplification and cross-lingual summarization. We empirically investigate the performance of SimCSum by comparing it with several strong baselines over several evaluation metrics and by human evaluation. Overall, SimCSum demonstrates statistically significant improvements over the state-of-the-art on two non-synthetic cross-lingual scientific datasets. Furthermore, we conduct an in-depth investigation into the linguistic properties of generated summaries and an error analysis.
Large Language Models (LLMs) have gathered significant attention due to their impressive performance on a variety of tasks. ChatGPT, developed by OpenAI, is a recent addition to the family of language models and is being called a disruptive technology by a few, owing to its human-like text-generation capabilities. Although, many anecdotal examples across the internet have evaluated ChatGPT's strength and weakness, only a few systematic research studies exist. To contribute to the body of literature of systematic research on ChatGPT, we evaluate the performance of ChatGPT on Abstractive Summarization by the means of automated metrics and blinded human reviewers. We also build automatic text classifiers to detect ChatGPT generated summaries. We found that while text classification algorithms can distinguish between real and generated summaries, humans are unable to distinguish between real summaries and those produced by ChatGPT.
The performance of text summarization has been greatly boosted by pre-trained language models. A main concern of existing methods is that most generated summaries are not factually inconsistent with their source documents. To alleviate the problem, many efforts have focused on developing effective factuality evaluation metrics based on natural language inference, question answering, and syntactic dependency et al. However, these approaches are limited by either their high computational complexity or the uncertainty introduced by multi-component pipelines, resulting in only partial agreement with human judgement. Most recently, large language models(LLMs) have shown excellent performance in not only text generation but also language comprehension. In this paper, we particularly explore ChatGPT's ability to evaluate factual inconsistency under a zero-shot setting by examining it on both coarse-grained and fine-grained evaluation tasks including binary entailment inference, summary ranking, and consistency rating. Experimental results indicate that ChatGPT generally outperforms previous evaluation metrics across the three tasks, indicating its great potential for factual inconsistency evaluation. However, a closer inspection of ChatGPT's output reveals certain limitations including its preference for more lexically similar candidates, false reasoning, and inadequate understanding of instructions.
Summarisation of research results in plain language is crucial for promoting public understanding of research findings. The use of Natural Language Processing to generate lay summaries has the potential to relieve researchers' workload and bridge the gap between science and society. The aim of this narrative literature review is to describe and compare the different text summarisation approaches used to generate lay summaries. We searched the databases Web of Science, Google Scholar, IEEE Xplore, Association for Computing Machinery Digital Library and arXiv for articles published until 6 May 2022. We included original studies on automatic text summarisation methods to generate lay summaries. We screened 82 articles and included eight relevant papers published between 2020 and 2021, all using the same dataset. The results show that transformer-based methods such as Bidirectional Encoder Representations from Transformers (BERT) and Pre-training with Extracted Gap-sentences for Abstractive Summarization (PEGASUS) dominate the landscape of lay text summarisation, with all but one study using these methods. A combination of extractive and abstractive summarisation methods in a hybrid approach was found to be most effective. Furthermore, pre-processing approaches to input text (e.g. applying extractive summarisation) or determining which sections of a text to include, appear critical. Evaluation metrics such as Recall-Oriented Understudy for Gisting Evaluation (ROUGE) were used, which do not consider readability. To conclude, automatic lay text summarisation is under-explored. Future research should consider long document lay text summarisation, including clinical trial reports, and the development of evaluation metrics that consider readability of the lay summary.
Sentence simplification tends to focus on the generic simplification of sentences by making them more readable and easier to understand. This paper provides a dataset aimed at training models that perform subject aware sentence simplifications rather than simplifying sentences as a whole. We also test models on that dataset which are inspired by model architecture used in abstractive summarization. We hand generated portions of the data and augment the dataset by further manipulating those hand written simplifications. Our results show that data-augmentation, data-masking, and model architecture choices used in summarization provide a solid baseline for comparison on subject aware simplification.
Neural abstractive summarization has been widely studied and achieved great success with large-scale corpora. However, the considerable cost of annotating data motivates the need for learning strategies under low-resource settings. In this paper, we investigate the problems of learning summarizers with only few examples and propose corresponding methods for improvements. First, typical transfer learning methods are prone to be affected by data properties and learning objectives in the pretext tasks. Therefore, based on pretrained language models, we further present a meta learning framework to transfer few-shot learning processes from source corpora to the target corpus. Second, previous methods learn from training examples without decomposing the content and preference. The generated summaries could therefore be constrained by the preference bias in the training set, especially under low-resource settings. As such, we propose decomposing the contents and preferences during learning through the parameter modulation, which enables control over preferences during inference. Third, given a target application, specifying required preferences could be non-trivial because the preferences may be difficult to derive through observations. Therefore, we propose a novel decoding method to automatically estimate suitable preferences and generate corresponding summary candidates from the few training examples. Extensive experiments demonstrate that our methods achieve state-of-the-art performance on six diverse corpora with 30.11%/33.95%/27.51% and 26.74%/31.14%/24.48% average improvements on ROUGE-1/2/L under 10- and 100-example settings.
People nowadays use search engines like Google, Yahoo, and Bing to find information on the Internet. Due to explosion in data, it is helpful for users if they are provided relevant summaries of the search results rather than just links to webpages. Text summarization has become a vital approach to help consumers swiftly grasp vast amounts of information.In this paper, different pre-trained models for text summarization are evaluated on different datasets. Specifically, we have used three different pre-trained models, namely, google/pegasus-cnn-dailymail, T5-base, facebook/bart-large-cnn. We have considered three different datasets, namely, CNN-dailymail, SAMSum and BillSum to get the output from the above three models. The pre-trained models are compared over these different datasets, each of 2000 examples, through ROUGH and BLEU metrics.