Task-oriented dialogue (TOD) systems aim to efficiently handle task-oriented conversations, including information collection. How to utilize TOD accurately, efficiently and effectively for information collection has always been a critical and challenging task. Recent studies have demonstrated that Large Language Models (LLMs) excel in dialogue, instruction generation, and reasoning, and can significantly enhance the performance of TOD through fine-tuning. However, current datasets primarily cater to user-led systems and are limited to predefined specific scenarios and slots, thereby necessitating improvements in the proactiveness, diversity, and capabilities of TOD. In this study, we present a detailed multi-domain task-oriented data construction process for conversations, and a Chinese dialogue dataset generated based on this process, TransferTOD, which authentically simulates human-computer dialogues in 30 popular life service scenarios. Leveraging this dataset, we trained a model called TransferTOD-7B using full-parameter fine-tuning, showcasing notable abilities in slot filling and questioning. Our work has demonstrated its strong generalization capabilities in various downstream scenarios, significantly enhancing both data utilization efficiency and system performance. The data is released in https://github.com/KongLongGeFDU/TransferTOD.

This paper presents the first study to explore the potential of parameter The remarkable performance of large language models (LLMs) has quantization for multimodal large language models to alleviate been well-established in recent literature [4, 9, 35, 36, 39], sparking the significant resource constraint encountered during visionlanguage a growing interest in the development of multimodal large language instruction tuning. We introduce a Quantization-aware models (MLLMs) [2, 3, 5, 24, 28, 32, 42]. This burgeoning field has Scale LeArning method based on multimodal Warmup, termed QS-led to substantial progress in a wide array of vision-language (VL) LAW. This method is grounded in two key innovations: (1) The tasks. To accomplish this, contemporary MLLMs primarily utilize learning of group-wise scale factors for quantized LLM weights multimodal instruction following examples for VL instruction tuning to mitigate the quantization error arising from activation outliers and adopt modular architectures [2, 21, 24, 28] to transform and achieve more effective vision-language instruction tuning; (2) visual features into the word embedding space of the LLM. This The implementation of a multimodal warmup that progressively innovative approach enables LLMs to execute multimodal tasks in integrates linguistic and multimodal training samples, thereby preventing an autoregressive fashion. One notable example of this technique is overfitting of the quantized model to multimodal data while LLaVA [24], which employs a linear projection layer to bridge the ensuring stable adaptation of multimodal large language models to gap between the visual encoder and the LLM.

Nobody knows how language works, but many theories abound. Transformers are a class of neural networks that process language automatically with more success than alternatives, both those based on neural computations and those that rely on other (e.g. more symbolic) mechanisms. Here, I highlight direct connections between the transformer architecture and certain theoretical perspectives on language. The empirical success of transformers relative to alternative models provides circumstantial evidence that the linguistic approaches that transformers embody should be, at least, evaluated with greater scrutiny by the linguistics community and, at best, considered to be the currently best available theories. Among the many unknown things about language is why we give specific names to particular objects or categories in the world around us. Brown (1958) discussed this question at length, noting that adult speakers of a language typically converge on the names that they give to common objects or categories. Building on this work, Rosch and Lloyd (1978) proposed that many semantic domains have clear basic level categories, such as apple, fish or knife (See Fig 1 (a) for an example).

Automatic extraction of procedural graphs from documents creates a low-cost way for users to easily understand a complex procedure by skimming visual graphs. Despite the progress in recent studies, it remains unanswered: whether the existing studies have well solved this task (Q1) and whether the emerging large language models (LLMs) can bring new opportunities to this task (Q2). To this end, we propose a new benchmark PAGED, equipped with a large high-quality dataset and standard evaluations. It investigates five state-of-the-art baselines, revealing that they fail to extract optimal procedural graphs well because of their heavy reliance on hand-written rules and limited available data. We further involve three advanced LLMs in PAGED and enhance them with a novel self-refine strategy. The results point out the advantages of LLMs in identifying textual elements and their gaps in building logical structures. We hope PAGED can serve as a major landmark for automatic procedural graph extraction and the investigations in PAGED can offer insights into the research on logic reasoning among non-sequential elements.

Active Learning (AL) allows models to learn interactively from user feedback. This paper introduces a counterfactual data augmentation approach to AL, particularly addressing the selection of datapoints for user querying, a pivotal concern in enhancing data efficiency. Our approach is inspired by Variation Theory, a theory of human concept learning that emphasizes the essential features of a concept by focusing on what stays the same and what changes. Instead of just querying with existing datapoints, our approach synthesizes artificial datapoints that highlight potential key similarities and differences among labels using a neuro-symbolic pipeline combining large language models (LLMs) and rule-based models. Through an experiment in the example domain of text classification, we show that our approach achieves significantly higher performance when there are fewer annotated data. As the annotated training data gets larger the impact of the generated data starts to diminish showing its capability to address the cold start problem in AL. This research sheds light on integrating theories of human learning into the optimization of AL.

In recent years, the application of Large Language Models (LLMs) in healthcare has shown significant promise in improving the accessibility and dissemination of medical knowledge. This paper presents a detailed study of various LLMs trained on the MedQuAD medical question-answering dataset, with a focus on identifying the most effective model for providing accurate medical information. Among the models tested, the Sentence-t5 combined with Mistral 7B demonstrated superior performance, achieving a precision score of 0.762. This model's enhanced capabilities are attributed to its advanced pretraining techniques, robust architecture, and effective prompt construction methodologies. By leveraging these strengths, the Sentence-t5 + Mistral 7B model excels in understanding and generating precise medical answers. Our findings highlight the potential of integrating sophisticated LLMs in medical contexts to facilitate efficient and accurate medical knowledge retrieval, thus significantly enhancing patient education and support.

Instruction tuning plays a critical role in aligning large language models (LLMs) with human preference. Despite the vast amount of open instruction datasets, naively training a LLM on all existing instructions may not be optimal and practical. To pinpoint the most beneficial datapoints, data assessment and selection methods have been proposed in the fields of natural language processing (NLP) and deep learning. However, under the context of instruction tuning, there still exists a gap in knowledge on what kind of data evaluation metrics can be employed and how they can be integrated into the selection mechanism. To bridge this gap, we present a comprehensive review on existing literature of data assessment and selection especially for instruction tuning of LLMs. We systematically categorize all applicable methods into quality-based, diversity-based, and importance-based ones where a unified, fine-grained taxonomy is structured. For each category, representative methods are elaborated to describe the landscape of relevant research. In addition, comparison between latest methods is conducted on their officially reported results to provide in-depth discussions on their limitations. Finally, we summarize the open challenges and propose the promosing avenues for future studies. All related contents are available at https://github.com/yuleiqin/fantastic-data-engineering.

The rapid evolution of software development necessitates enhanced safety measures. Extracting information about software failures from companies is becoming increasingly more available through news articles. This research utilizes the Failure Analysis Investigation with LLMs (FAIL) model to extract industry-specific information. Although the FAIL model's database is rich in information, it could benefit from further categorization and industry-specific insights to further assist software engineers. In previous work news articles were collected from reputable sources and categorized by incidents inside a database. Prompt engineering and Large Language Models (LLMs) were then applied to extract relevant information regarding the software failure. This research extends these methods by categorizing articles into specific domains and types of software failures. The results are visually represented through graphs. The analysis shows that throughout the database some software failures occur significantly more often in specific industries. This categorization provides a valuable resource for software engineers and companies to identify and address common failures. This research highlights the synergy between software engineering and Large Language Models (LLMs) to automate and enhance the analysis of software failures. By transforming data from the database into an industry specific model, we provide a valuable resource that can be used to identify common vulnerabilities, predict potential risks, and implement proactive measures for preventing software failures. Leveraging the power of the current FAIL database and data visualization, we aim to provide an avenue for safer and more secure software in the future.

Large language models (LLMs) can help writers build story worlds by generating world elements, such as factions, characters, and locations. However, making sense of many generated elements can be overwhelming. Moreover, if the user wants to precisely control aspects of generated elements that are difficult to specify verbally, prompting alone may be insufficient. We introduce Patchview, a customizable LLM-powered system that visually aids worldbuilding by allowing users to interact with story concepts and elements through the physical metaphor of magnets and dust. Elements in Patchview are visually dragged closer to concepts with high relevance, facilitating sensemaking. The user can also steer the generation with verbally elusive concepts by indicating the desired position of the element between concepts. When the user disagrees with the LLM's visualization and generation, they can correct those by repositioning the element. These corrections can be used to align the LLM's future behaviors to the user's perception. With a user study, we show that Patchview supports the sensemaking of world elements and steering of element generation, facilitating exploration during the worldbuilding process. Patchview provides insights on how customizable visual representation can help sensemake, steer, and align generative AI model behaviors with the user's intentions.

The utilization of automated depression detection significantly enhances early intervention for individuals experiencing depression. Despite numerous proposals on automated depression detection using recorded clinical interview videos, limited attention has been paid to considering the hierarchical structure of the interview questions. In clinical interviews for diagnosing depression, clinicians use a structured questionnaire that includes routine baseline questions and follow-up questions to assess the interviewee's condition. This paper introduces HiQuE (Hierarchical Question Embedding network), a novel depression detection framework that leverages the hierarchical relationship between primary and follow-up questions in clinical interviews. HiQuE can effectively capture the importance of each question in diagnosing depression by learning mutual information across multiple modalities. We conduct extensive experiments on the widely-used clinical interview data, DAIC-WOZ, where our model outperforms other state-of-the-art multimodal depression detection models and emotion recognition models, showcasing its clinical utility in depression detection.