Detecting salient parts in text using natural language processing has been widely used to mitigate the effects of information overflow. Nevertheless, most of the datasets available for this task are derived mainly from academic publications. We introduce SPACE-IDEAS, a dataset for salient information detection from innovation ideas related to the Space domain. The text in SPACE-IDEAS varies greatly and includes informal, technical, academic and business-oriented writing styles. In addition to a manually annotated dataset we release an extended version that is annotated using a large generative language model. We train different sentence and sequential sentence classifiers, and show that the automatically annotated dataset can be leveraged using multitask learning to train better classifiers.
Quickly detecting problematic research articles is crucial to safeguarding the integrity of scientific research. This study explores whether Twitter mentions of retracted articles can signal potential problems with the articles prior to their retraction, potentially serving as an early warning system for scholars. To investigate this, we analysed a dataset of 4,354 Twitter mentions associated with 504 retracted articles. The effectiveness of Twitter mentions in predicting article retractions was evaluated by both manual and Large Language Model (LLM) labelling. Manual labelling results indicated that 25.7% of tweets signalled problems before retraction. Using the manual labelling results as the baseline, we found that LLMs (GPT-4o-mini, Gemini 1.5 Flash, and Claude-3.5-Haiku) outperformed lexicon-based sentiment analysis tools (e.g., TextBlob) in detecting potential problems, suggesting that automatic detection of problematic articles from social media using LLMs is technically feasible. Nevertheless, since only a small proportion of retracted articles (11.1%) were criticised on Twitter prior to retraction, such automatic systems would detect only a minority of problematic articles. Overall, this study offers insights into how social media data, coupled with emerging generative AI techniques, can support research integrity.
Significant methodological strides have been made toward Chest X-ray (CXR) understanding via modern vision-language models (VLMs), demonstrating impressive Visual Question Answering (VQA) and CXR report generation abilities. However, existing CXR understanding frameworks still possess several procedural caveats. (1) Previous methods solely use CXR reports, which are insufficient for comprehensive Visual Question Answering (VQA), especially when additional health-related data like medication history and prior diagnoses are needed. (2) Previous methods use raw CXR reports, which are often arbitrarily structured. While modern language models can understand various text formats, restructuring reports for clearer, organized anatomy-based information could enhance their usefulness. (3) Current evaluation methods for CXR-VQA primarily emphasize linguistic correctness, lacking the capability to offer nuanced assessments of the generated answers. In this work, to address the aforementioned caveats, we introduce WoLF, a Wide-scope Large Language Model Framework for CXR understanding. To resolve (1), we capture multi-faceted records of patients, which are utilized for accurate diagnoses in real-world clinical scenarios. Specifically, we adopt the Electronic Health Records (EHR) to generate instruction-following data suited for CXR understanding. Regarding (2), we enhance report generation performance by decoupling knowledge in CXR reports based on anatomical structure even within the attention step via masked attention. To address (3), we introduce an AI-evaluation protocol optimized for assessing the capabilities of LLM. Through extensive experimental validations, WoLF demonstrates superior performance over other models on MIMIC-CXR in the AI-evaluation arena about VQA (up to +9.47%p mean score) and by metrics about report generation (+7.3%p BLEU-1).
Language models (LMs) are powerful yet mostly for text generation tasks. Tools have substantially enhanced their performance for tasks that require complex skills. However, many works adopt the term "tool" in different ways, raising the question: What is a tool anyway? Subsequently, where and how do tools help LMs? In this survey, we provide a unified definition of tools as external programs used by LMs, and perform a systematic review of LM tooling scenarios and approaches. Grounded on this review, we empirically study the efficiency of various tooling methods by measuring their required compute and performance gains on various benchmarks, and highlight some challenges and potential future research in the field.
Existing works focus on fixed-size layout pattern generation, while the more practical free-size pattern generation receives limited attention. In this paper, we propose ChatPattern, a novel Large-Language-Model (LLM) powered framework for flexible pattern customization. ChatPattern utilizes a two-part system featuring an expert LLM agent and a highly controllable layout pattern generator. The LLM agent can interpret natural language requirements and operate design tools to meet specified needs, while the generator excels in conditional layout generation, pattern modification, and memory-friendly patterns extension. Experiments on challenging pattern generation setting shows the ability of ChatPattern to synthesize high-quality large-scale patterns.
Large language models (LLMs) have attracted significant attention for potential applications in digital health, while their application in mental health is subject to ongoing debate. This systematic review aims to evaluate the usage of LLMs in mental health, focusing on their strengths and limitations in early screening, digital interventions, and clinical applications. Adhering to PRISMA guidelines, we searched PubMed, IEEE Xplore, Scopus, JMIR, and ACM using keywords: 'mental health OR mental illness OR mental disorder OR psychiatry' AND 'large language models'. We included articles published between January 1, 2017, and April 30, 2024, excluding non-English articles. 30 articles were evaluated, which included research on mental health conditions and suicidal ideation detection through text (n=15), usage of LLMs for mental health conversational agents (CAs) (n=7), and other applications and evaluations of LLMs in mental health (n=18). LLMs exhibit substantial effectiveness in detecting mental health issues and providing accessible, de-stigmatized eHealth services. However, the current risks associated with the clinical use might surpass their benefits. The study identifies several significant issues: the lack of multilingual datasets annotated by experts, concerns about the accuracy and reliability of the content generated, challenges in interpretability due to the 'black box' nature of LLMs, and persistent ethical dilemmas. These include the lack of a clear ethical framework, concerns about data privacy, and the potential for over-reliance on LLMs by both therapists and patients, which could compromise traditional medical practice. Despite these issues, the rapid development of LLMs underscores their potential as new clinical aids, emphasizing the need for continued research and development in this area.
In traditional decision making processes, social biases of human decision makers can lead to unequal economic outcomes for underrepresented social groups, such as women, racial or ethnic minorities. Recently, the increasing popularity of Large language model based artificial intelligence suggests a potential transition from human to AI based decision making. How would this impact the distributional outcomes across social groups? Here we investigate the gender and racial biases of OpenAIs GPT, a widely used LLM, in a high stakes decision making setting, specifically assessing entry level job candidates from diverse social groups. Instructing GPT to score approximately 361000 resumes with randomized social identities, we find that the LLM awards higher assessment scores for female candidates with similar work experience, education, and skills, while lower scores for black male candidates with comparable qualifications. These biases may result in a 1 or 2 percentage point difference in hiring probabilities for otherwise similar candidates at a certain threshold and are consistent across various job positions and subsamples. Meanwhile, we also find stronger pro female and weaker anti black male patterns in democratic states. Our results demonstrate that this LLM based AI system has the potential to mitigate the gender bias, but it may not necessarily cure the racial bias. Further research is needed to comprehend the root causes of these outcomes and develop strategies to minimize the remaining biases in AI systems. As AI based decision making tools are increasingly employed across diverse domains, our findings underscore the necessity of understanding and addressing the potential unequal outcomes to ensure equitable outcomes across social groups.
In this project, we examine how freelancers changed their strategic positioning on an online work platform following the launch of ChatGPT in November 2022 - a major advance in AI technologies. We document that post-ChatGPT, freelancers bid on fewer jobs and reposition themselves by differentiating their distribution of bids (i.e., job applications) relative to their prior behavior. We disentangle heterogeneity in strategic responses by exploring how exposure to changes in demand or supply shape incumbent repositioning. We find that the launch of ChatGPT was associated with a short-term decrease in labor demand and an increase in labor supply, though these changes vary across work domains. In response to decreases in labor demand, workers changed their horizontal positioning and withdrew from the platform. In response to increases in labor supply, workers were less likely to decrease bidding or reposition horizontally but shifted their vertical position by targeting lower-value jobs. We further show that repositioning is less likely for high-skill freelancers who face greater adjustment costs. This research contributes to our understanding of how and why workers respond to technological change in the context of recent advances in AI technologies.
Fact-checking is the task of verifying the factuality of a given claim by examining the available evidence. High-quality evidence plays a vital role in enhancing fact-checking systems and facilitating the generation of explanations that are understandable to humans. However, the provision of both sufficient and relevant evidence for explainable fact-checking systems poses a challenge. To tackle this challenge, we propose a method based on a Large Language Model to automatically retrieve and summarize evidence from the Web. Furthermore, we construct RU22Fact, a novel multilingual explainable fact-checking dataset on the Russia-Ukraine conflict in 2022 of 16K samples, each containing real-world claims, optimized evidence, and referenced explanation. To establish a baseline for our dataset, we also develop an end-to-end explainable fact-checking system to verify claims and generate explanations. Experimental results demonstrate the prospect of optimized evidence in increasing fact-checking performance and also indicate the possibility of further progress in the end-to-end claim verification and explanation generation tasks.
In this paper, we borrow the large language model (LLM) ChatGPT-3.5 to automatically and quickly design a new metaheuristic algorithm (MA) with only a small amount of input. The novel animal-inspired MA named zoological search optimization (ZSO) draws inspiration from the collective behaviors of animals for solving continuous optimization problems. Specifically, the basic ZSO algorithm involves two search operators: the prey-predator interaction operator and the social flocking operator to balance exploration and exploitation well. Besides, the standard prompt engineering framework CRISPE (i.e., Capacity and Role, Insight, Statement, Personality, and Experiment) is responsible for the specific prompt design. Furthermore, we designed four variants of the ZSO algorithm with slight human-interacted adjustment. In numerical experiments, we comprehensively investigate the performance of ZSO-derived algorithms on CEC2014 benchmark functions, CEC2022 benchmark functions, and six engineering optimization problems. 20 popular and state-of-the-art MAs are employed as competitors. The experimental results and statistical analysis confirm the efficiency and effectiveness of ZSO-derived algorithms. At the end of this paper, we explore the prospects for the development of the metaheuristics community under the LLM era.
The rise of large language models (LLMs) has opened new opportunities in Recommender Systems (RSs) by enhancing user behavior modeling and content understanding. However, current approaches that integrate LLMs into RSs solely utilize either LLM or conventional recommender model (CRM) to generate final recommendations, without considering which data segments LLM or CRM excel in. To fill in this gap, we conduct experiments on MovieLens-1M and Amazon-Books datasets, and compare the performance of a representative CRM (DCNv2) and an LLM (LLaMA2-7B) on various groups of data samples. Our findings reveal that LLMs excel in data segments where CRMs exhibit lower confidence and precision, while samples where CRM excels are relatively challenging for LLM, requiring substantial training data and a long training time for comparable performance. This suggests potential synergies in the combination between LLM and CRM. Motivated by these insights, we propose Collaborative Recommendation with conventional Recommender and Large Language Model (dubbed \textit{CoReLLa}). In this framework, we first jointly train LLM and CRM and address the issue of decision boundary shifts through alignment loss. Then, the resource-efficient CRM, with a shorter inference time, handles simple and moderate samples, while LLM processes the small subset of challenging samples for CRM. Our experimental results demonstrate that CoReLLa outperforms state-of-the-art CRM and LLM methods significantly, underscoring its effectiveness in recommendation tasks.
Performance of large language models (LLMs) may vary with different prompts or instructions of even the same task. One commonly recognized factor for this phenomenon is the model's familiarity with the given prompt or instruction, which is typically estimated by its perplexity. However, finding the prompt with the lowest perplexity is challenging, given the enormous space of possible prompting phrases. In this paper, we propose monotonic paraphrasing (MonoPara), an end-to-end decoding strategy that paraphrases given prompts or instructions into their lower perplexity counterparts based on an ensemble of a paraphrase LM for prompt (or instruction) rewriting, and a target LM (i.e. the prompt or instruction executor) that constrains the generation for lower perplexity. The ensemble decoding process can efficiently paraphrase the original prompt without altering its semantic meaning, while monotonically decreasing the perplexity of each generation as calculated by the target LM. We explore in detail both greedy and search-based decoding as two alternative decoding schemes of MonoPara. Notably, MonoPara does not require any training and can monotonically lower the perplexity of the paraphrased prompt or instruction, leading to improved performance of zero-shot LM prompting as evaluated on a wide selection of tasks. In addition, MonoPara is also shown to effectively improve LMs' generalization on perturbed and unseen task instructions.
The recent advancements in artificial intelligence highlight the potential of language models in psychological health support. While models trained on data from mental health service platform have achieved preliminary success, challenges persist in areas such as data scarcity, quality, and ensuring a solid foundation in psychological techniques. To address these challenges, this study introduces a novel approach to enhance the precision and efficacy of psychological support through large language models. Specifically, we design a specific prompt derived from principles of Cognitive Behavioral Therapy (CBT) and have generated the CBT QA dataset, specifically for Chinese psychological health Q&A based on CBT structured intervention strategies. Unlike previous methods, our dataset emphasizes professional and structured response. Utilizing this dataset, we fine-tuned the large language model, giving birth to CBT-LLM, the large-scale language model specifically designed for Cognitive Behavioral Therapy techniques. Empirical evaluations demonstrate that CBT-LLM excels in generating structured, professional, and highly relevant responses in psychological health support tasks, showcasing its practicality and quality. The model is available on Hugging Face: https://huggingface.co/Hongbin37/CBT-LLM.
A major public concern regarding the training of large language models (LLMs) is whether they abusing copyrighted online text. Previous membership inference methods may be misled by similar examples in vast amounts of training data. Additionally, these methods are often too complex for general users to understand and use, making them centralized, lacking transparency, and trustworthiness. To address these issues, we propose an alternative \textit{insert-and-detection} methodology, advocating that web users and content platforms employ \textbf{\textit{unique identifiers}} for reliable and independent membership inference. Users and platforms can create their own identifiers, embed them in copyrighted text, and independently detect them in future LLMs. As an initial demonstration, we introduce \textit{ghost sentences}, a primitive form of unique identifiers, consisting primarily of passphrases made up of random words. By embedding one ghost sentences in a few copyrighted texts, users can detect its membership using a perplexity test and a \textit{user-friendly} last-$k$ words test. The perplexity test is based on the fact that LLMs trained on natural language should exhibit high perplexity when encountering unnatural passphrases. As the repetition increases, users can leverage the verbatim memorization ability of LLMs to perform a last-$k$ words test by chatting with LLMs without writing any code. Both tests offer rigorous statistical guarantees for membership inference. For LLaMA-13B, a perplexity test on 30 ghost sentences with an average of 7 repetitions in 148K examples yields a 0.891 ROC AUC. For the last-$k$ words test with OpenLLaMA-3B, 11 out of 16 users, with an average of 24 examples each, successfully identify their data from 1.8M examples.
The capabilities of AI for biomedicine span a wide spectrum, from the atomic level, where it solves partial differential equations for quantum systems, to the molecular level, predicting chemical or protein structures, and further extending to societal predictions like infectious disease outbreaks. Recent advancements in large language models, exemplified by models like ChatGPT, have showcased significant prowess in natural language tasks, such as translating languages, constructing chatbots, and answering questions. When we consider biomedical data, we observe a resemblance to natural language in terms of sequences: biomedical literature and health records presented as text, biological sequences or sequencing data arranged in sequences, or sensor data like brain signals as time series. The question arises: Can we harness the potential of recent large language models to drive biomedical knowledge discoveries? In this survey, we will explore the application of large language models to three crucial categories of biomedical data: 1) textual data, 2) biological sequences, and 3) brain signals. Furthermore, we will delve into large language model challenges in biomedical research, including ensuring trustworthiness, achieving personalization, and adapting to multi-modal data representation
Autocomplete suggestions are fundamental to modern text entry systems, with applications in domains such as messaging and email composition. Typically, autocomplete suggestions are generated from a language model with a confidence threshold. However, this threshold does not directly take into account the cognitive load imposed on the user by surfacing suggestions, such as the effort to switch contexts from typing to reading the suggestion, and the time to decide whether to accept the suggestion. In this paper, we study the problem of improving inline autocomplete suggestions in text entry systems via a sequential decision-making formulation, and use reinforcement learning to learn suggestion policies through repeated interactions with a target user over time. This formulation allows us to factor cognitive load into the objective of training an autocomplete model, through a reward function based on text entry speed. We acquired theoretical and experimental evidence that, under certain objectives, the sequential decision-making formulation of the autocomplete problem provides a better suggestion policy than myopic single-step reasoning. However, aligning these objectives with real users requires further exploration. In particular, we hypothesize that the objectives under which sequential decision-making can improve autocomplete systems are not tailored solely to text entry speed, but more broadly to metrics such as user satisfaction and convenience.
The advancement of Large Language Models (LLMs) has significantly transformed the field of natural language processing, although the focus on English-centric models has created a noticeable research gap for specific languages, including Vietnamese. To address this issue, this paper presents vi-mistral-x, an innovative Large Language Model designed expressly for the Vietnamese language. It utilizes a unique method of continual pre-training, based on the Mistral architecture, which incorporates grouped-query attention and sliding window attention techniques. This model, vi-Mistral-X, marks a significant step forward in improving the understanding and generation of the Vietnamese language. It introduces an additional phase of continual pre-training, specifically adapted for Vietnamese, enhancing the model's capability in understanding complex language nuances and generating accurate, context-aware Vietnamese text. Through comprehensive testing on various benchmarks, vi-mistral-x has shown to outperform existing Vietnamese LLMs in several key areas, including text classification, question answering, and text generation. Particularly, in the Vietnamese Multitask Language Understanding (VMLU) benchmark, vi-mistral-x sets a new standard, outperforming other available models significantly. This paper highlights the critical role of continual pre-training in advancing language-specific LLMs and opens new avenues for the development of multilingual models. We aim for vi-mistral-x to not just be an important asset for processing the Vietnamese language but also to encourage more advancements in creating large language models for languages that are less represented.
Graph self-supervised learning (SSL) is now a go-to method for pre-training graph foundation models (GFMs). There is a wide variety of knowledge patterns embedded in the graph data, such as node properties and clusters, which are crucial to learning generalized representations for GFMs. However, existing surveys of GFMs have several shortcomings: they lack comprehensiveness regarding the most recent progress, have unclear categorization of self-supervised methods, and take a limited architecture-based perspective that is restricted to only certain types of graph models. As the ultimate goal of GFMs is to learn generalized graph knowledge, we provide a comprehensive survey of self-supervised GFMs from a novel knowledge-based perspective. We propose a knowledge-based taxonomy, which categorizes self-supervised graph models by the specific graph knowledge utilized. Our taxonomy consists of microscopic (nodes, links, etc.), mesoscopic (context, clusters, etc.), and macroscopic knowledge (global structure, manifolds, etc.). It covers a total of 9 knowledge categories and more than 25 pretext tasks for pre-training GFMs, as well as various downstream task generalization strategies. Such a knowledge-based taxonomy allows us to re-examine graph models based on new architectures more clearly, such as graph language models, as well as provide more in-depth insights for constructing GFMs.
We introduce Gaussian masking for Language-Image Pre-Training (GLIP) a novel, straightforward, and effective technique for masking image patches during pre-training of a vision-language model. GLIP builds on Fast Language-Image Pre-Training (FLIP), which randomly masks image patches while training a CLIP model. GLIP replaces random masking with centered masking, that uses a Gaussian distribution and is inspired by the importance of image patches at the center of the image. GLIP retains the same computational savings as FLIP, while improving performance across a range of downstream datasets and tasks, as demonstrated by our experimental results. We show the benefits of GLIP to be easy to obtain, requiring no delicate tuning of the Gaussian, and also applicable to data sets containing images without an obvious center focus.
Recent works often assume that Vision-Language Model (VLM) representations are based on visual attributes like shape. However, it is unclear to what extent VLMs prioritize this information to represent concepts. We propose Extract and Explore (EX2), a novel approach to characterize textual features that are important for VLMs. EX2 uses reinforcement learning to align a large language model with VLM preferences and generates descriptions that incorporate features that are important for the VLM. Then, we inspect the descriptions to identify features that contribute to VLM representations. Using EX2, we find that spurious descriptions have a major role in VLM representations despite providing no helpful information, e.g., Click to enlarge photo of CONCEPT. More importantly, among informative descriptions, VLMs rely significantly on non-visual attributes like habitat (e.g., North America) to represent visual concepts. Also, our analysis reveals that different VLMs prioritize different attributes in their representations. Overall, we show that VLMs do not simply match images to scene descriptions and that non-visual or even spurious descriptions significantly influence their representations.