We present Adjacent Possible Exploration (APE), a selective fine-tuning method for adapting large language models that systematically explores parameter modifications while maintaining model stability. Inspired by evolutionary optimization principles, APE evaluates multiple candidate parameter updates through fine-tuning on small data subsets and accepts only those exceeding a performance threshold. Unlike standard fine-tuning that follows single gradient directions, APE implements a filtered selection process that prevents destabilizing parameter changes while enabling systematic improvement. Our method achieves 33.9\% BLEU improvement and 36.2\% perplexity reduction on news summarization tasks while using minimal computational resources. The approach provides a practical framework for controlled model adaptation that balances performance gains with representational stability.
News data have become an essential resource across various disciplines, including economics, finance, management, social sciences, and computer science. Researchers leverage newspaper articles to study economic trends, market dynamics, corporate strategies, public perception, political discourse, and the evolution of public opinion. Additionally, news datasets have been instrumental in training large-scale language models, with applications in sentiment analysis, fake news detection, and automated news summarization. Despite their significance, access to comprehensive news corpora remains a key challenge. Many full-text news providers, such as Factiva and LexisNexis, require costly subscriptions, while free alternatives often suffer from incomplete data and transparency issues. This paper presents a novel approach to obtaining full-text newspaper articles at near-zero cost by leveraging data from the Global Database of Events, Language, and Tone (GDELT). Specifically, we focus on the GDELT Web News NGrams 3.0 dataset, which provides high-frequency updates of n-grams extracted from global online news sources. We provide Python code to reconstruct full-text articles from these n-grams by identifying overlapping textual fragments and intelligently merging them. Our method enables researchers to access structured, large-scale newspaper data for text analysis while overcoming the limitations of existing proprietary datasets. The proposed approach enhances the accessibility of news data for empirical research, facilitating applications in economic forecasting, computational social science, and natural language processing.
The automation of news analysis and summarization presents a promising solution to the challenge of processing and analyzing vast amounts of information prevalent in today's information society. Large Language Models (LLMs) have demonstrated the capability to transform vast amounts of textual data into concise and easily comprehensible summaries, offering an effective solution to the problem of information overload and providing users with a quick overview of relevant information. A particularly significant application of this technology lies in supply chain risk analysis. Companies must monitor the news about their suppliers and respond to incidents for several critical reasons, including compliance with laws and regulations, risk management, and maintaining supply chain resilience. This paper develops an automated news summarization system for supply chain risk analysis using LLMs. The proposed solution aggregates news from various sources, summarizes them using LLMs, and presents the condensed information to users in a clear and concise format. This approach enables companies to optimize their information processing and make informed decisions. Our study addresses two main research questions: (1) Are LLMs effective in automating news summarization, particularly in the context of supply chain risk analysis? (2) How effective are various LLMs in terms of readability, duplicate detection, and risk identification in their summarization quality? In this paper, we conducted an offline study using a range of publicly available LLMs at the time and complemented it with a user study focused on the top performing systems of the offline experiments to evaluate their effectiveness further. Our results demonstrate that LLMs, particularly Few-Shot GPT-4o mini, offer significant improvements in summary quality and risk identification.
The increasing demand for efficient summarization tools in resource-constrained environments highlights the need for effective solutions. While large language models (LLMs) deliver superior summarization quality, their high computational resource requirements limit practical use applications. In contrast, small language models (SLMs) present a more accessible alternative, capable of real-time summarization on edge devices. However, their summarization capabilities and comparative performance against LLMs remain underexplored. This paper addresses this gap by presenting a comprehensive evaluation of 19 SLMs for news summarization across 2,000 news samples, focusing on relevance, coherence, factual consistency, and summary length. Our findings reveal significant variations in SLM performance, with top-performing models such as Phi3-Mini and Llama3.2-3B-Ins achieving results comparable to those of 70B LLMs while generating more concise summaries. Notably, SLMs are better suited for simple prompts, as overly complex prompts may lead to a decline in summary quality. Additionally, our analysis indicates that instruction tuning does not consistently enhance the news summarization capabilities of SLMs. This research not only contributes to the understanding of SLMs but also provides practical insights for researchers seeking efficient summarization solutions that balance performance and resource use.
In the Learning to Defer (L2D) framework, a prediction model can either make a prediction or defer it to an expert, as determined by a rejector. Current L2D methods train the rejector to decide whether to reject the entire prediction, which is not desirable when the model predicts long sequences. We present an L2D setting for sequence outputs where the system can defer specific outputs of the whole model prediction to an expert in an effort to interleave the expert and machine throughout the prediction. We propose two types of model-based post-hoc rejectors for pre-trained predictors: a token-level rejector, which defers specific token predictions to experts with next token prediction capabilities, and a one-time rejector for experts without such abilities, which defers the remaining sequence from a specific point onward. In the experiments, we also empirically demonstrate that such granular deferrals achieve better cost-accuracy tradeoffs than whole deferrals on Traveling salesman solvers and News summarization models.
Given the recent introduction of multiple language models and the ongoing demand for improved Natural Language Processing tasks, particularly summarization, this work provides a comprehensive benchmarking of 20 recent language models, focusing on smaller ones for the news summarization task. In this work, we systematically test the capabilities and effectiveness of these models in summarizing news article texts which are written in different styles and presented in three distinct datasets. Specifically, we focus in this study on zero-shot and few-shot learning settings and we apply a robust evaluation methodology that combines different evaluation concepts including automatic metrics, human evaluation, and LLM-as-a-judge. Interestingly, including demonstration examples in the few-shot learning setting did not enhance models' performance and, in some cases, even led to worse quality of the generated summaries. This issue arises mainly due to the poor quality of the gold summaries that have been used as reference summaries, which negatively impacts the models' performance. Furthermore, our study's results highlight the exceptional performance of GPT-3.5-Turbo and GPT-4, which generally dominate due to their advanced capabilities. However, among the public models evaluated, certain models such as Qwen1.5-7B, SOLAR-10.7B-Instruct-v1.0, Meta-Llama-3-8B and Zephyr-7B-Beta demonstrated promising results. These models showed significant potential, positioning them as competitive alternatives to large models for the task of news summarization.
Recent large language models (LLMs) have demonstrated a remarkable ability to perform natural language understanding and generation tasks. In this work, we investigate the use of LLMs for evaluating faithfulness in news summarization, finding that it achieves a strong correlation with human judgments. We further investigate LLMs' capabilities as a faithfulness post-editor, experimenting with different chain-of-thought prompts for locating and correcting factual inconsistencies between a generated summary and the source news document and are able to achieve a higher editing success rate than was reported in prior work. We perform both automated and human evaluations of the post-edited summaries, finding that prompting LLMs using chain-of-thought reasoning about factual error types is an effective faithfulness post-editing strategy, performing comparably to fine-tuned post-editing models. We also demonstrate that multiple rounds of post-editing, which has not previously been explored, can be used to gradually improve the faithfulness of summaries whose errors cannot be fully corrected in a single round.
Explainable AI (XAI) techniques for text summarization provide valuable understanding of how the summaries are generated. Recent studies have highlighted a major challenge in this area, known as the disagreement problem. This problem occurs when different XAI methods offer contradictory explanations for the summary generated from the same input article. This inconsistency across XAI methods has been evaluated using predefined metrics designed to quantify agreement levels between them, revealing significant disagreement. This impedes the reliability and interpretability of XAI in this area. To address this challenge, we propose a novel approach that utilizes sentence transformers and the k-means clustering algorithm to first segment the input article and then generate the explanation of the summary generated for each segment. By producing regional or segmented explanations rather than comprehensive ones, a decrease in the observed disagreement between XAI methods is hypothesized. This segmentation-based approach was used on two news summarization datasets, namely Extreme Summarization(XSum) and CNN-DailyMail, and the experiment was conducted using multiple disagreement metrics. Our experiments validate the hypothesis by showing a significant reduction in disagreement among different XAI methods. Additionally, a JavaScript visualization tool is developed, that is easy to use and allows users to interactively explore the color-coded visualization of the input article and the machine-generated summary based on the attribution scores of each sentences.
Existing research on news summarization primarily focuses on single-language single-document (SLSD), single-language multi-document (SLMD) or cross-language single-document (CLSD). However, in real-world scenarios, news about a international event often involves multiple documents in different languages, i.e., mixed-language multi-document (MLMD). Therefore, summarizing MLMD news is of great significance. However, the lack of datasets for MLMD news summarization has constrained the development of research in this area. To fill this gap, we construct a mixed-language multi-document news summarization dataset (MLMD-news), which contains four different languages and 10,992 source document cluster and target summary pairs. Additionally, we propose a graph-based extract-generate model and benchmark various methods on the MLMD-news dataset and publicly release our dataset and code\footnote[1]{https://github.com/Southnf9/MLMD-news}, aiming to advance research in summarization within MLMD scenarios.
News summarization in today's global scene can be daunting with its flood of multilingual content and varied viewpoints from different sources. However, current studies often neglect such real-world scenarios as they tend to focus solely on either single-language or single-document tasks. To bridge this gap, we aim to unify Multi-lingual, Cross-lingual and Multi-document Summarization into a novel task, i.e., MCMS, which encapsulates the real-world requirements all-in-one. Nevertheless, the lack of a benchmark inhibits researchers from adequately studying this invaluable problem. To tackle this, we have meticulously constructed the GLOBESUMM dataset by first collecting a wealth of multilingual news reports and restructuring them into event-centric format. Additionally, we introduce the method of protocol-guided prompting for high-quality and cost-effective reference annotation. In MCMS, we also highlight the challenge of conflicts between news reports, in addition to the issues of redundancies and omissions, further enhancing the complexity of GLOBESUMM. Through extensive experimental analysis, we validate the quality of our dataset and elucidate the inherent challenges of the task. We firmly believe that GLOBESUMM, given its challenging nature, will greatly contribute to the multilingual communities and the evaluation of LLMs.
Large Language Models (LLM) are increasingly trained on data generated by other LLM, either because generated text and images become part of the pre-training corpus, or because synthetized data is used as a replacement for expensive human-annotation. This raises concerns about \emph{model collapse}, a drop in model performance when their training sets include generated data. Considering that it is easier for both humans and machines to tell between good and bad examples than to generate high-quality samples, we investigate the use of verification on synthesized data to prevent model collapse. We provide a theoretical characterization using Gaussian mixtures, linear classifiers, and linear verifiers to derive conditions with measurable proxies to assess whether the verifier can effectively select synthesized data that leads to optimal performance. We experiment with two practical tasks -- computing matrix eigenvalues with transformers and news summarization with LLMs -- which both exhibit model collapse when trained on generated data, and show that verifiers, even imperfect ones, can indeed be harnessed to prevent model collapse and that our proposed proxy measure strongly correlates with performance.
With the deluge of information delivered by the daily news cycle, there is a growing need to effectively and efficiently summarize news feeds for quick consumption. We leverage large language models (LLMs), with their advanced learning and generative abilities as compared to conventional language models, to generate concise and coherent summaries for news articles from the XSum dataset. Our paper focuses on two key aspects of LLMs: Efficient in-context Learning (ELearn) and Parameter Efficient Fine-tuning (EFit). Under ELearn, we find that increasing the number of shots in prompts and utilizing simple templates generally improve the quality of summaries. We also find that utilizing relevant examples in few-shot learning for ELearn does not improve model performance. In addition, we studied EFit using different methods and demonstrate that fine-tuning the first layer of LLMs produces better outcomes as compared to fine-tuning other layers or utilizing LoRA. We also find that leveraging more relevant training samples using selective layers does not result in better performance. By combining ELearn and EFit, we create a new model (ELearnFit) that leverages the benefits of both few-shot learning and fine-tuning and produces superior performance to either model alone. We also use ELearnFit to highlight the trade-offs between prompting and fine-tuning, especially for situations where only a limited number of annotated samples are available. Ultimately, our research provides practical techniques to optimize news summarization during the prompting and fine-tuning stages and enhances the synthesis of news articles.
Text summarization models have typically focused on optimizing aspects of quality such as fluency, relevance, and coherence, particularly in the context of news articles. However, summarization models are increasingly being used to summarize diverse sources of text, such as social media data, that encompass a wide demographic user base. It is thus crucial to assess not only the quality of the generated summaries, but also the extent to which they can fairly represent the opinions of diverse social groups. Position bias, a long-known issue in news summarization, has received limited attention in the context of social multi-document summarization. We deeply investigate this phenomenon by analyzing the effect of group ordering in input documents when summarizing tweets from three distinct linguistic communities: African-American English, Hispanic-aligned Language, and White-aligned Language. Our empirical analysis shows that although the textual quality of the summaries remains consistent regardless of the input document order, in terms of fairness, the results vary significantly depending on how the dialect groups are presented in the input data. Our results suggest that position bias manifests differently in social multi-document summarization, severely impacting the fairness of summarization models.
The rapid adoption of Electronic Health Records (EHRs) has been instrumental in streamlining administrative tasks, increasing transparency, and enabling continuity of care across providers. An unintended consequence of the increased documentation burden, however, has been reduced face-time with patients and, concomitantly, a dramatic rise in clinician burnout. In this thesis, we pinpoint a particularly time-intensive, yet critical, documentation task: generating a summary of a patient's hospital admissions, and propose and evaluate automated solutions. In Chapter 2, we construct a dataset based on 109,000 hospitalizations (2M source notes) and perform exploratory analyses to motivate future work on modeling and evaluation [NAACL 2021]. In Chapter 3, we address faithfulness from a modeling perspective by revising noisy references [EMNLP 2022] and, to reduce the reliance on references, directly calibrating model outputs to metrics [ACL 2023]. These works relied heavily on automatic metrics as human annotations were limited. To fill this gap, in Chapter 4, we conduct a fine-grained expert annotation of system errors in order to meta-evaluate existing metrics and better understand task-specific issues of domain adaptation and source-summary alignments. To learn a metric less correlated to extractiveness (copy-and-paste), we derive noisy faithfulness labels from an ensemble of existing metrics and train a faithfulness classifier on these pseudo labels [MLHC 2023]. Finally, in Chapter 5, we demonstrate that fine-tuned LLMs (Mistral and Zephyr) are highly prone to entity hallucinations and cover fewer salient entities. We improve both coverage and faithfulness by performing sentence-level entity planning based on a set of pre-computed salient entities from the source text, which extends our work on entity-guided news summarization [ACL, 2023], [EMNLP, 2023].
The wide adoption of Large language models (LLMs) makes their dependability a pressing concern. Detection of errors is the first step to mitigating their impact on a system and thus, efficient error detection for LLMs is an important issue. In many settings, the LLM is considered as a black box with no access to the internal nodes; this prevents the use of many error detection schemes that need access to the model's internal nodes. An interesting observation is that the output of LLMs in error-free operation should be valid and normal text. Therefore, when the text is not valid or differs significantly from normal text, it is likely that there is an error. Based on this observation we propose to perform Concurrent Linguistic Error Detection (CLED); this scheme extracts some linguistic features of the text generated by the LLM and feeds them to a concurrent classifier that detects errors. Since the proposed error detection mechanism only relies on the outputs of the model, then it can be used on LLMs in which there is no access to the internal nodes. The proposed CLED scheme has been evaluated on the T5 model when used for news summarization and on the OPUS-MT model when used for translation. In both cases, the same set of linguistic features has been used for error detection to illustrate the applicability of the proposed scheme beyond a specific case. The results show that CLED can detect most of the errors at a low overhead penalty. The use of the concurrent classifier also enables a trade-off between error detection effectiveness and its associated overhead, so providing flexibility to a designer.
The advent of Large Language Models (LLMs) has led to remarkable progress on a wide range of natural language processing tasks. Despite the advances, these large-sized models still suffer from hallucinating information in their output, which poses a major issue in automatic text summarization, as we must guarantee that the generated summary is consistent with the content of the source document. Previous research addresses the challenging task of detecting hallucinations in the output (i.e. inconsistency detection) in order to evaluate the faithfulness of the generated summaries. However, these works primarily focus on English and recent multilingual approaches lack German data. This work presents absinth, a manually annotated dataset for hallucination detection in German news summarization and explores the capabilities of novel open-source LLMs on this task in both fine-tuning and in-context learning settings. We open-source and release the absinth dataset to foster further research on hallucination detection in German.
Modern instruction-tuned models have become highly capable in text generation tasks such as summarization, and are expected to be released at a steady pace. In practice one may now wish to choose confidently, but with minimal effort, the best performing summarization model when applied to a new domain or purpose. In this work, we empirically investigate the test sample size necessary to select a preferred model in the context of news summarization. Empirical results reveal that comparative evaluation converges quickly for both automatic and human evaluation, with clear preferences for a system emerging from under 100 examples. The human preference data allows us to quantify how well automatic scores can reproduce preference rankings across a variety of downstream summarization tasks. We find that, while automatic metrics are stable at smaller sample sizes, only some automatic metrics are able to moderately predict model win rates according to human preference.
Single document news summarization has seen substantial progress on faithfulness in recent years, driven by research on the evaluation of factual consistency, or hallucinations. We ask whether these advances carry over to other text summarization domains. We propose a new evaluation benchmark on topic-focused dialogue summarization, generated by LLMs of varying sizes. We provide binary sentence-level human annotations of the factual consistency of these summaries along with detailed explanations of factually inconsistent sentences. Our analysis shows that existing LLMs hallucinate significant amounts of factual errors in the dialogue domain, regardless of the model's size. On the other hand, when LLMs, including GPT-4, serve as binary factual evaluators, they perform poorly and can be outperformed by prevailing state-of-the-art specialized factuality evaluation metrics. Finally, we conducted an analysis of hallucination types with a curated error taxonomy. We find that there are diverse errors and error distributions in model-generated summaries and that non-LLM based metrics can capture all error types better than LLM-based evaluators.
Query-focused summarization (QFS) aims to provide a summary of a single document/multi documents that can satisfy the information needs of a given query. It is useful for various real-world applications, such as abstractive snippet generation or more recent retrieval augmented generation (RAG). A prototypical QFS pipeline consists of a retriever (sparse or dense retrieval) and a generator (usually a large language model). However, applying large language models (LLM) potentially leads to hallucinations, especially when the evidence contradicts the prior belief of LLMs. There has been growing interest in developing new decoding methods to improve generation quality and reduce hallucination. In this work, we conduct a large-scale reproducibility study on one recently proposed decoding method -- Context-aware Decoding (CAD). In addition to replicating CAD's experiments on news summarization datasets, we include experiments on QFS datasets, and conduct more rigorous analysis on computational complexity and hyperparameter sensitivity. Experiments with eight different language models show that performance-wise, CAD improves QFS quality by (1) reducing factuality errors/hallucinations while (2) mostly retaining the match of lexical patterns, measured by ROUGE scores, while also at a cost of increased inference-time FLOPs and reduced decoding speed. The code implementation based on Huggingface Library is made available https://github.com/zhichaoxu-shufe/context-aware-decoding-qfs
Multi-document summarization is the process of automatically generating a concise summary of multiple documents related to the same topic. This summary can help users quickly understand the key information from a large collection of documents. Multi-document summarization systems are more complex than single-document summarization systems due to the need to identify and combine information from multiple sources. In this paper, we have developed a machine learning model that generates a concise summary of a topic from multiple news documents. The model is designed to be unbiased by sampling its input equally from all the different aspects of the topic, even if the majority of the news sources lean one way.