With more and more advanced data analysis techniques emerging, people will expect these techniques to be applied in more complex tasks and solve problems in our daily lives. Text Summarization is one of famous applications in Natural Language Processing (NLP) field. It aims to automatically generate summary with important information based on a given context, which is important when you have to deal with piles of documents. Summarization techniques can help capture key points in a short time and bring convenience in works. One of applicable situation is meeting summarization, especially for important meeting that tend to be long, complicated, multi-topic and multi-person. Therefore, when people want to review specific content from a meeting, it will be hard and time-consuming to find the related spans in the meeting transcript. However, most of previous works focus on doing summarization for newsletters, scientific articles...etc, which have a clear document structure and an official format. For the documents with complex structure like transcripts, we think those works are not quite suitable for meeting summarization. Besides, the consistency of summary is another issue common to be discussed in NLP field. To conquer challenges of meeting summarization, we are inspired by "QMSum: A New Benchmark for Query-based Multi-domain Meeting Summarization" proposed by Microsoft and we also propose our Locater model designed to extract relevant spans based on given transcript and query, which are then summarized by Summarizer model. Furthermore, we perform a comparative study by applying different word embedding techniques to improve summary consistency.
This paper presents TL;DR Progress, a new tool for exploring the literature on neural text summarization. It organizes 514~papers based on a comprehensive annotation scheme for text summarization approaches and enables fine-grained, faceted search. Each paper was manually annotated to capture aspects such as evaluation metrics, quality dimensions, learning paradigms, challenges addressed, datasets, and document domains. In addition, a succinct indicative summary is provided for each paper, consisting of automatically extracted contextual factors, issues, and proposed solutions. The tool is available online at https://www.tldr-progress.de, a demo video at https://youtu.be/uCVRGFvXUj8
The rapid dissemination of misinformation through social media increased the importance of automated fact-checking. Furthermore, studies on what deep neural models pay attention to when making predictions have increased in recent years. While significant progress has been made in this field, it has not yet reached a level of reasoning comparable to human reasoning. To address these gaps, we propose a multi-task explainable neural model for misinformation detection. Specifically, this work formulates an explanation generation process of the model's veracity prediction as a text summarization problem. Additionally, the performance of the proposed model is discussed on publicly available datasets and the findings are evaluated with related studies.
In this paper, we propose Ranksum, an approach for extractive text summarization of single documents based on the rank fusion of four multi-dimensional sentence features extracted for each sentence: topic information, semantic content, significant keywords, and position. The Ranksum obtains the sentence saliency rankings corresponding to each feature in an unsupervised way followed by the weighted fusion of the four scores to rank the sentences according to their significance. The scores are generated in completely unsupervised way, and a labeled document set is required to learn the fusion weights. Since we found that the fusion weights can generalize to other datasets, we consider the Ranksum as an unsupervised approach. To determine topic rank, we employ probabilistic topic models whereas semantic information is captured using sentence embeddings. To derive rankings using sentence embeddings, we utilize Siamese networks to produce abstractive sentence representation and then we formulate a novel strategy to arrange them in their order of importance. A graph-based strategy is applied to find the significant keywords and related sentence rankings in the document. We also formulate a sentence novelty measure based on bigrams, trigrams, and sentence embeddings to eliminate redundant sentences from the summary. The ranks of all the sentences computed for each feature are finally fused to get the final score for each sentence in the document. We evaluate our approach on publicly available summarization datasets CNN/DailyMail and DUC 2002. Experimental results show that our approach outperforms other existing state-of-the-art summarization methods.
With the abundance of data and information in todays time, it is nearly impossible for man, or, even machine, to go through all of the data line by line. What one usually does is to try to skim through the lines and retain the absolutely important information, that in a more formal term is called summarization. Text summarization is an important task that aims to compress lengthy documents or articles into shorter, coherent representations while preserving the core information and meaning. This project introduces an innovative approach to text summarization, leveraging the capabilities of Graph Neural Networks (GNNs) and Named Entity Recognition (NER) systems. GNNs, with their exceptional ability to capture and process the relational data inherent in textual information, are adept at understanding the complex structures within large documents. Meanwhile, NER systems contribute by identifying and emphasizing key entities, ensuring that the summarization process maintains a focus on the most critical aspects of the text. By integrating these two technologies, our method aims to enhances the efficiency of summarization and also tries to ensures a high degree relevance in the condensed content. This project, therefore, offers a promising direction for handling the ever increasing volume of textual data in an information-saturated world.
Personalized large language models (LLMs) are designed to tailor responses to individual user preferences. While Reinforcement Learning from Human Feedback (RLHF) is a commonly used framework for aligning LLMs with human preferences, vanilla RLHF assumes that all human preferences share the same distribution, preventing fine-tuned LLMs from generating personalized content when user preferences are diverse. In this work, we propose Personalized-RLHF (P-RLHF), an efficient framework that utilizes a lightweight user model to capture individual user preferences and jointly learns the user model and the personalized LLM from human feedback. P-RLHF exhibits the following three characteristics: (1) It enables an LLM to generate personalized content and scale efficiently with growing number of users. (2) It handles both explicit user preferences described as textual input and implicit user preferences encoded in the feedback data. (3) It eliminates the need for users to fully articulate their preferences, which are normally needed for prompting LLMs to generate personalized content yet are often impractical to obtain in real-world scenarios. Our experimental results show that personalized LLMs trained using P-RLHF generate responses that are more closely aligned with individual user preferences, outperforming vanilla, non-personalized RLHF and prompting-based personalization approaches across different tasks. We opensource our code at https://github.com/HumainLab/Personalized_RLHF.
YouTube faces a global crisis with the dissemination of false information and hate speech. To counter these issues, YouTube has implemented strict rules against uploading content that includes false information or promotes hate speech. While numerous studies have been conducted to reduce offensive English-language content, there's a significant lack of research on Sinhala content. This study aims to address the aforementioned gap by proposing a solution to minimize the spread of violence and misinformation in Sinhala YouTube videos. The approach involves developing a rating system that assesses whether a video contains false information by comparing the title and description with the audio content and evaluating whether the video includes hate speech. The methodology encompasses several steps, including audio extraction using the Pytube library, audio transcription via the fine-tuned Whisper model, hate speech detection employing the distilroberta-base model and a text classification LSTM model, and text summarization through the fine-tuned BART-Large- XSUM model. Notably, the Whisper model achieved a 48.99\% word error rate, while the distilroberta-base model demonstrated an F1 score of 0.856 and a recall value of 0.861 in comparison to the LSTM model, which exhibited signs of overfitting.
Large Language Models (LLMs) have shown remarkable capabilities across a wide variety of Natural Language Processing (NLP) tasks and have attracted attention from multiple domains, including financial services. Despite the extensive research into general-domain LLMs, and their immense potential in finance, Financial LLM (FinLLM) research remains limited. This survey provides a comprehensive overview of FinLLMs, including their history, techniques, performance, and opportunities and challenges. Firstly, we present a chronological overview of general-domain Pre-trained Language Models (PLMs) through to current FinLLMs, including the GPT-series, selected open-source LLMs, and financial LMs. Secondly, we compare five techniques used across financial PLMs and FinLLMs, including training methods, training data, and fine-tuning methods. Thirdly, we summarize the performance evaluations of six benchmark tasks and datasets. In addition, we provide eight advanced financial NLP tasks and datasets for developing more sophisticated FinLLMs. Finally, we discuss the opportunities and the challenges facing FinLLMs, such as hallucination, privacy, and efficiency. To support AI research in finance, we compile a collection of accessible datasets and evaluation benchmarks on GitHub.
Text summarization is the process of condensing a piece of text to fewer sentences, while still preserving its content. Chat transcript, in this context, is a textual copy of a digital or online conversation between a customer (caller) and agent(s). This paper presents an indigenously (locally) developed hybrid method that first combines extractive and abstractive summarization techniques in compressing ill-punctuated or un-punctuated chat transcripts to produce more readable punctuated summaries and then optimizes the overall quality of summarization through reinforcement learning. Extensive testing, evaluations, comparisons, and validation have demonstrated the efficacy of this approach for large-scale deployment of chat transcript summarization, in the absence of manually generated reference (annotated) summaries.
The impressive performance of large language models (LLMs) has attracted considerable attention from the academic and industrial communities. Besides how to construct and train LLMs, how to effectively evaluate and compare the capacity of LLMs has also been well recognized as an important yet difficult problem. Existing paradigms rely on either human annotators or model-based evaluators to evaluate the performance of LLMs on different tasks. However, these paradigms often suffer from high cost, low generalizability, and inherited biases in practice, which make them incapable of supporting the sustainable development of LLMs in long term. In order to address these issues, inspired by the peer review systems widely used in academic publication process, we propose a novel framework that can automatically evaluate LLMs through a peer-review process. Specifically, for the evaluation of a specific task, we first construct a small qualification exam to select "reviewers" from a couple of powerful LLMs. Then, to actually evaluate the "submissions" written by different candidate LLMs, i.e., the evaluatees, we use the reviewer LLMs to rate or compare the submissions. The final ranking of evaluatee LLMs is generated based on the results provided by all reviewers. We conducted extensive experiments on text summarization tasks with eleven LLMs including GPT-4. The results demonstrate the existence of biasness when evaluating using a single LLM. Also, our PRE model outperforms all the baselines, illustrating the effectiveness of the peer review mechanism.
We present the Radiation Oncology NLP Database (ROND), the first dedicated Natural Language Processing (NLP) dataset for radiation oncology, an important medical specialty that has received limited attention from the NLP community in the past. With the advent of Artificial General Intelligence (AGI), there is an increasing need for specialized datasets and benchmarks to facilitate research and development. ROND is specifically designed to address this gap in the domain of radiation oncology, a field that offers many opportunities for NLP exploration. It encompasses various NLP tasks including Logic Reasoning, Text Classification, Named Entity Recognition (NER), Question Answering (QA), Text Summarization, and Patient-Clinician Conversations, each with a distinct focus on radiation oncology concepts and application cases. In addition, we have developed an instruction-tuning dataset consisting of over 20k instruction pairs (based on ROND) and trained a large language model, CancerChat. This serves to demonstrate the potential of instruction-tuning large language models within a highly-specialized medical domain. The evaluation results in this study could serve as baseline results for future research. ROND aims to stimulate advancements in radiation oncology and clinical NLP by offering a platform for testing and improving algorithms and models in a domain-specific context. The ROND dataset is a joint effort of multiple U.S. health institutions. The data is available at https://github.com/zl-liu/Radiation-Oncology-NLP-Database.
Generative large language models (LLMs) have demonstrated exceptional proficiency in various natural language processing (NLP) tasks, including machine translation, question answering, text summarization, and natural language understanding. To further enhance the performance of LLMs in machine translation, we conducted an investigation into two popular prompting methods and their combination, focusing on cross-language combinations of Persian, English, and Russian. We employed n-shot feeding and tailored prompting frameworks. Our findings indicate that multilingual LLMs like PaLM exhibit human-like machine translation outputs, enabling superior fine-tuning of desired translation nuances in accordance with style guidelines and linguistic considerations. These models also excel in processing and applying prompts. However, the choice of language model, machine translation task, and the specific source and target languages necessitate certain considerations when adopting prompting frameworks and utilizing n-shot in-context learning. Furthermore, we identified errors and limitations inherent in popular LLMs as machine translation tools and categorized them based on various linguistic metrics. This typology of errors provides valuable insights for utilizing LLMs effectively and offers methods for designing prompts for in-context learning. Our report aims to contribute to the advancement of machine translation with LLMs by improving both the accuracy and reliability of evaluation metrics.
Minimum Bayes-Risk (MBR) decoding is shown to be a powerful alternative to beam search decoding for a wide range of text generation tasks. However, MBR requires a huge amount of time for inference to compute the MBR objective, which makes the method infeasible in many situations where response time is critical. Confidence-based pruning (CBP) (Cheng and Vlachos, 2023) has recently been proposed to reduce the inference time in machine translation tasks. Although it is shown to significantly reduce the amount of computation, it requires hyperparameter tuning using a development set to be effective. To this end, we propose Approximate Minimum Bayes-Risk (AMBR) decoding, a hyperparameter-free method to run MBR decoding approximately. AMBR is derived from the observation that the problem of computing the sample-based MBR objective is the medoid identification problem. AMBR uses the Correlated Sequential Halving (CSH) algorithm (Baharav and Tse, 2019), the best approximation algorithm to date for the medoid identification problem, to compute the sample-based MBR objective. We evaluate AMBR on machine translation, text summarization, and image captioning tasks. The results show that AMBR achieves on par with CBP, with CBP selecting hyperparameters through an Oracle for each given computation budget.
Despite the widespread use of the Persian language by millions globally, limited efforts have been made in natural language processing for this language. The use of large language models as effective tools in various natural language processing tasks typically requires extensive textual data and robust hardware resources. Consequently, the scarcity of Persian textual data and the unavailability of powerful hardware resources have hindered the development of large language models for Persian. This paper introduces the first large Persian language model, named PersianLLaMA, trained on a collection of Persian texts and datasets. This foundational model comes in two versions, with 7 and 13 billion parameters, trained on formal and colloquial Persian texts using two different approaches. PersianLLaMA has been evaluated for natural language generation tasks based on the latest evaluation methods, namely using larger language models, and for natural language understanding tasks based on automated machine metrics. The results indicate that PersianLLaMA significantly outperforms its competitors in both understanding and generating Persian text. PersianLLaMA marks an important step in the development of Persian natural language processing and can be a valuable resource for the Persian-speaking community. This large language model can be used for various natural language processing tasks, especially text generation like chatbots, question-answering, machine translation, and text summarization
Large Language Models (LLMs) are adept at text manipulation -- tasks such as machine translation and text summarization. However, these models can also be prone to hallucination, which can be detrimental to the faithfulness of any answers that the model provides. Recent works in combating hallucinations in LLMs deal with identifying hallucinated sentences and categorizing the different ways in which models hallucinate. This paper takes a deep dive into LLM behavior with respect to hallucinations, defines a token-level approach to identifying different kinds of hallucinations, and further utilizes this token-level tagging to improve the interpretability and faithfulness of LLMs in dialogue summarization tasks. Through this, the paper presents a new, enhanced dataset and a new training paradigm.
As Large Language Models (LLMs) have made significant advancements across various tasks, such as question answering, translation, text summarization, and dialogue systems, the need for accuracy in information becomes crucial, especially for serious financial products serving billions of users like Alipay. However, for a real-world product serving millions of users, the inference speed of LLMs becomes a critical factor compared to a mere experimental model. Hence, this paper presents a generic framework for accelerating the inference process, resulting in a substantial increase in speed and cost reduction for our LLM-based scenarios, with lossless generation accuracy. In the traditional inference process, each token is generated sequentially by the LLM, leading to a time consumption proportional to the number of generated tokens. To enhance this process, our framework, named \textit{lookahead}, introduces a \textit{multi-branch} strategy. Instead of generating a single token at a time, we propose a Trie-based retrieval and verification mechanism to be able to accept several tokens at a forward step. Our strategy offers two distinct advantages: (1) it guarantees absolute correctness of the output, avoiding any approximation algorithms, and (2) the worst-case performance of our approach is equivalent to the conventional process. We conduct extensive experiments to demonstrate the significant improvements achieved by applying our inference acceleration framework. Our framework is widely deployed in Alipay since April 2023, and obtain remarkable 2.66x to 6.26x speedup. Our code is available at https://github.com/alipay/PainlessInferenceAcceleration.
Modern models for text generation show state-of-the-art results in many natural language processing tasks. In this work, we explore the effectiveness of abstractive text summarization models for keyphrase selection. A list of keyphrases is an important element of a text in databases and repositories of electronic documents. In our experiments, abstractive text summarization models fine-tuned for keyphrase generation show quite high results for a target text corpus. However, in most cases, the zero-shot performance on other corpora and domains is significantly lower. We investigate cross-domain limitations of abstractive text summarization models for keyphrase generation. We present an evaluation of the fine-tuned BART models for the keyphrase selection task across six benchmark corpora for keyphrase extraction including scientific texts from two domains and news texts. We explore the role of transfer learning between different domains to improve the BART model performance on small text corpora. Our experiments show that preliminary fine-tuning on out-of-domain corpora can be effective under conditions of a limited number of samples.
Abstractive text summarization is surging with the number of training samples to cater to the needs of the deep learning models. These models tend to exploit the training data representations to attain superior performance by improving the quantitative element of the resultant summary. However, increasing the size of the training set may not always be the ideal solution to maximize the performance, and therefore, a need to revisit the quality of training samples and the learning protocol of deep learning models is a must. In this paper, we aim to discretize the vector space of the abstractive text summarization models to understand the characteristics learned between the input embedding space and the models' encoder space. We show that deep models fail to capture the diversity of the input space. Further, the distribution of data points on the encoder space indicates that an unchecked increase in the training samples does not add value; rather, a tear-down of data samples is highly needed to make the models focus on variability and faithfulness. We employ clustering techniques to learn the diversity of a model's sample space and how data points are mapped from the embedding space to the encoder space and vice versa. Further, we devise a metric to filter out redundant data points to make the model more robust and less data hungry. We benchmark our proposed method using quantitative metrics, such as Rouge, and qualitative metrics, such as BERTScore, FEQA and Pyramid score. We also quantify the reasons that inhibit the models from learning the diversity from the varied input samples.
The pursuit of article or text summarization has captured the attention of natural language processing (NLP) practitioners, presenting itself as a formidable challenge. ChatGPT 3.5 exhibits the capacity to condense the content of up to 3000 tokens into a single page, aiming to retain pivotal information from a given text across diverse themes. In a conducted qualitative research endeavor, we selected seven scientific articles and employed the publicly available ChatGPT service to generate summaries of these articles. Subsequently, we engaged six co-authors of the articles in a survey, presenting five questions to evaluate the quality of the summaries compared to the original content. The findings revealed that the summaries produced by ChatGPT effectively encapsulated the crucial information present in the articles, preserving the principal message of each manuscript. Nonetheless, there was a slight diminishment in the technical depth of the summaries as opposed to the original articles. As a result, our conclusion underscores ChatGPT's text summarization capability as a potent tool for extracting essential insights in a manner more aligned with reporting than purely scientific discourse.
Reinforcement learning from human feedback (RLHF) has emerged as the main paradigm for aligning large language models (LLMs) with human preferences. Typically, RLHF involves the initial step of learning a reward model from human feedback, often expressed as preferences between pairs of text generations produced by a pre-trained LLM. Subsequently, the LLM's policy is fine-tuned by optimizing it to maximize the reward model through a reinforcement learning algorithm. However, an inherent limitation of current reward models is their inability to fully represent the richness of human preferences and their dependency on the sampling distribution. In this study, we introduce an alternative pipeline for the fine-tuning of LLMs using pairwise human feedback. Our approach entails the initial learning of a preference model, which is conditioned on two inputs given a prompt, followed by the pursuit of a policy that consistently generates responses preferred over those generated by any competing policy, thus defining the Nash equilibrium of this preference model. We term this approach Nash learning from human feedback (NLHF). In the context of a tabular policy representation, we present a novel algorithmic solution, Nash-MD, founded on the principles of mirror descent. This algorithm produces a sequence of policies, with the last iteration converging to the regularized Nash equilibrium. Additionally, we explore parametric representations of policies and introduce gradient descent algorithms for deep-learning architectures. To demonstrate the effectiveness of our approach, we present experimental results involving the fine-tuning of a LLM for a text summarization task. We believe NLHF offers a compelling avenue for preference learning and policy optimization with the potential of advancing the field of aligning LLMs with human preferences.