检索结果-内蒙古大学图书馆

Teaching the Pre-trained Model to Generate Simple Texts for Text Simplification

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Sun, Renliang Xu, Wei Wan, Xiaojun Wangxuan Institute of Computer Technology Peking University China Center for Data Science Peking University China The MOE Key Laboratory of Computational Linguistics Peking University China School of Interactive Computing Georgia Institute of Technology United States

Randomly masking text spans in ordinary texts in the pre-training stage hardly allows models to acquire the ability to generate simple texts. It can hurt the performance of pre-trained models on text simplification tasks. In this paper, we propose a new continued pre-training strategy to teach the pre-trained model to generate simple texts. We continue pre-training BART, a representative model, to obtain SimpleBART. It consistently and significantly improves the results on lexical simplification, sentence simplification, and document-level simplification tasks over BART. At the end, we compare SimpleBART with several representative large language models (LLMs). © 2023, CC BY-SA.

关键词： Computational linguistics

Visual-Imagery-Based Analogical Construction in Geometric Matrix Reasoning Task

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Yang, Yuan McGreggor, Keith Kunda, Maithilee Electrical Engineering and Computer Science Vanderbilt University NashvilleTN37235 United States School of Interactive Computing Georgia Tech AtlantaGA30308 United States

Analogical reasoning fundamentally involves exploiting redundancy in a given task, but there are various strategies for an intelligent agent to identify and exploit such redundancy, often resulting in very different levels of reasoning ability. We explore such variations of analogy in geometric reasoning task, namely the Raven’s Progressive Matrices. We show how different analogical constructions used by the same basic imagery-based computational model — varying only in how they "slice" a matrix problem into parts and search within/across these parts — achieve very different test scores, substantially overlapping the range of human performance. Our findings suggest that the ability to build effective high-level analogical constructions is as important as competencies in low-level reasoning, which raises interesting questions about the extent to which building the "right" analogies contributes to individual differences in human reasoning and how intelligent agents might learn to build among different constructions in the first place. © 2022, CC BY.

关键词： Redundancy

SURE-VQA: SYSTEMATIC UNDERSTANDING OF ROBUSTNESS EVALUATION IN MEDICAL VQA TASKS

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Kahl, Kim-Celine Erkan, Selen Traub, Jeremias Lüth, Carsten T. Maier-Hein, Klaus Maier-Hein, Lena Jaeger, Paul F. Heidelberg Interactive Machine Learning Group Germany Heidelberg Germany Faculty of Mathematics and Computer Science University of Heidelberg Germany Heidelberg Division of Medical Image Computing Germany Germany Pattern Analysis and Learning Group Department of Radiation Oncology Heidelberg University Hospital Germany Heidelberg Germany

Vision-Language Models (VLMs) have great potential in medical tasks, like Visual Question Answering (VQA), where they could act as interactive assistants for both patients and clinicians. Yet their robustness to distribution shifts on unseen data remains a critical concern for safe deployment. Evaluating such robustness requires a controlled experimental setup that allows for systematic insights into the model's behavior. However, we demonstrate that current setups fail to offer sufficiently thorough evaluations, limiting their ability to accurately assess model robustness. To address this gap, our work introduces a novel framework, called SURE-VQA, centered around three key requirements to overcome the current pitfalls and systematically analyze the robustness of VLMs: 1) Since robustness on synthetic shifts does not necessarily translate to real-world shifts, robustness should be measured on real-world shifts that are inherent to the VQA data;2) Traditional token-matching metrics often fail to capture underlying semantics, necessitating the use of large language models (LLMs) for more accurate semantic evaluation;3) Model performance often lacks interpretability due to missing sanity baselines, thus meaningful baselines should be reported that allow assessing the multimodal impact on the VLM. To demonstrate the relevance of this framework, we conduct a study on the robustness of various Fine-Tuning methods across three medical datasets with four different types of distribution shifts. Our study reveals several important findings: 1) Sanity baselines that do not utilize image data can perform surprisingly well;2) We confirm LoRA as the best-performing PEFT method;3) No PEFT method consistently outperforms others in terms of robustness to shifts. Code is provided at https://***/IML-DKFZ/sure-vqa. Copyright © 2024, The Authors. All rights reserved.

关键词： Semantics

A Computational Interface to Translate Strategic Intent from Unstructured Language in a Low-Data Setting

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Tambwekar, Pradyumna Dodeja, Lakshita Vaska, Nathan Xu, Wei Gombolay, Matthew School of Interactive Computing Georgia Institute of Technology United States Computer Science Department Brown University United States Massachusetts Institute of Technology Lincoln Laboratory United States

Many real-world tasks involve a mixed-initiative setup, wherein humans and AI systems collaboratively perform a task. While significant work has been conducted towards enabling humans to specify, through language, exactly how an agent should complete a task (i.e., low-level specification), prior work lacks on interpreting the high-level strategic intent of the human commanders. Parsing strategic intent from language will allow autonomous systems to independently operate according to the user’s plan without frequent guidance or instruction. In this paper, we build a computational interface capable of translating unstructured language strategies into actionable intent in the form of goals and constraints. Leveraging a game environment, we collect a dataset of over 1000 examples, mapping language strategies to the corresponding goals and constraints, and show that our model, trained on this dataset, significantly outperforms human interpreters in inferring strategic intent (i.e., goals and constraints) from language (p Copyright © 2022, The Authors. All rights reserved.

关键词： High level languages

Neural semi-Markov CRF for monolingual word alignment

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Lan, Wuwei Jiang, Chao Xu, Wei Department of Computer Science and Engineering Ohio State University School of Interactive Computing Georgia Institute of Technology

Monolingual word alignment is important for studying fine-grained editing operations (i.e., deletion, addition, and substitution) in text-to-text generation tasks, such as paraphrase generation, text simplification, neutralizing biased language, etc. In this paper, we present a novel neural semi-Markov CRF alignment model, which unifies word and phrase alignments through variable-length spans. We also create a new benchmark with human annotations that cover four different text genres to evaluate monolingual word alignment models in more realistic settings. Experimental results show that our proposed model outperforms all previous approaches for monolingual word alignment as well as a competitive QA-based baseline, which was previously only applied to bilingual data. Our model demonstrates good generalizability to three out-of-domain datasets and shows great utility in two downstream applications: automatic text simplification and sentence pair classification tasks.1 Copyright © 2021, The Authors. All rights reserved.

关键词： Alignment

This is the Table I Want! interactive Data Transformation on Desktop and in Virtual Reality

学校读者我要写书评

暂无评论

arXiv 2023年

作者： In, Sungwon Lin, Tica North, Chris Pfister, Hanspeter Yang, Yalong The Department of Computer Science Virginia Tech BlacksburgVA24060 United States The School of Interactive Computing Georgia Institute of Technology AtlantaGA30309 United States The Department of Computer Science Virginia Tech United States John A. Paulson School of Engineering and Applied Sciences Harvard University CambridgeMA02138 United States

Data transformation is an essential step in data science. While experts primarily use programming to transform their data, there is an increasing need to support non-programmers with user interface-based tools. With the rapid development in interaction techniques and computing environments, we report our empirical findings about the effects of interaction techniques and environments on performing data transformation tasks. Specifically, we studied the potential benefits of direct interaction and virtual reality (VR) for data transformation. We compared gesture interaction versus a standard WIMP user interface, each on the desktop and in VR. With the tested data and tasks, we found time performance was similar between desktop and VR. Meanwhile, VR demonstrates preliminary evidence to better support provenance and sense-making throughout the data transformation process. Our exploration of performing data transformation in VR also provides initial affirmation for enabling an iterative and fully immersive data science workflow. © 2023, CC BY.

关键词： Virtual reality

Deep Video Representation Learning: A Survey

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Ravanbakhsh, Elham Liang, Yongqing Ramanujam, J. Li, Xin Division of Electrical & Computer Engineering Center for Computation & Technology Louisiana State University Baton RougeLA70803 United States Department of Computer Science and Engineering Texas A&M University College StationTX77843 United States Section of Visual Computing and Interactive Media Texas A&M University College StationTX77843 United States

This paper provides a review on representation learning for videos. We classify recent spatio-temporal feature learning methods for sequential visual data and compare their pros and cons for general video analysis. Building effective features for videos is a fundamental problem in computer vision tasks involving video analysis and understanding. Existing features can be generally categorized into spatial and temporal features. Their effectiveness under variations of illumination, occlusion, view and background are discussed. Finally, we discuss the remaining challenges in existing deep video representation learning studies. © 2024, CC BY-NC-ND.

关键词： Deep learning

Enhancing Weakly Supervised Semantic Segmentation with Multi-modal Foundation Models: An End-to-End Approach

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Ravanbakhsh, Elham Niu, Cheng Liang, Yongqing Ramanujam, J. Li, Xin Louisiana State University Baton RougeLA70803 United States Department of Computer Science & Engineering Texas A&M University College StationTX77843 United States Section of Visual Computing and Interactive Media Texas A&M University College StationTX77843 United States

Semantic segmentation is a core computer vision problem, but the high costs of data annotation have hindered its wide application. Weakly-Supervised Semantic Segmentation (WSSS) offers a cost-efficient workaround to extensive labeling in comparison to fully-supervised methods by using partial or incomplete labels. Existing WSSS methods have difficulties in learning the boundaries of objects leading to poor segmentation results. We propose a novel and effective framework that addresses these issues by leveraging visual foundation models inside the bounding box. Adopting a two-stage WSSS framework, our proposed network consists of a pseudo-label generation module and a segmentation module. The first stage leverages Segment Anything Model (SAM) to generate high-quality pseudo-labels. To alleviate the problem of delineating precise boundaries, we adopt SAM inside the bounding box with the help of another pre-trained foundation model (e.g., Grounding-DINO). Furthermore, we eliminate the necessity of using the supervision of image labels, by employing CLIP in classification. Then in the second stage, the generated high-quality pseudo-labels are used to train an off-the-shelf segmenter that achieves the state-of-the-art performance on PASCAL VOC 2012 and MS COCO 2014. Copyright © 2024, The Authors. All rights reserved.

关键词： Semantic Segmentation

A Toolbox of Feminist Wonder: Theories and methods that can make a difference 26

学校读者我要写书评

暂无评论

A Toolbox of Feminist Wonder: Theories and methods that can ...

26th ACM Conference on computer-Supported Cooperative Work and Social computing, CSCW 2023

作者： Hansson, Karin Bardzell, Shaowen Bhandari, Aparajita Boulicault, Marion Doyle, Dylan Thomas Erete, Sheena Cerratto Pargman, Teresa Lazem, Shaimaa Muller, Michael Normark, Maria Petterson, Adrian Rosén, Anton Poikolainen Taylor, Alex S. Thomas, Jakita O. Watson, Julia Södertörn University The School of Natural Sciences Technology and Environmental Studies Sweden Georgia Institute of Technology The School of Interactive Computing United States Cornell University Department of Communication United States University of Edinburgh Department of Philosophy United States University of Colorado Boulder Department of Information Science United States University of Maryland College of Information United States Stockholm University The Department of Computer and Systems Sciences Sweden City for Scientific Research and Technological Applications The Informatics Research Institute Egypt Ibm Research Human Centered Ai United States Uppsala University Department of Informatics and Media Sweden University of Toronto Faculty of Information Canada Aalto University The School of Arts Design and Architecture Finland City University of London The Centre for Human Computer Interaction Design United Kingdom Auburn University Department of Computer Science and Software Engineering United States University of Toronto Department of Computer Science Canada

ISBN: (纸本)9798400701290

This one-day hybrid workshop builds on previous feminist CSCW workshops to explore feminist theoretical and methodological approaches that have provided us with useful tools to see things differently and make space for change. Since its inception over a decade ago, feminist HCI has progressed from the margins to mainstream HCI, with numerous references in the literature. Feminist HCI has also evolved to incorporate other critical HCI practices such as Queer HCI, participatory design, and speculative design. While feminist approaches have grown in popularity and become mainstream, it is getting more difficult to distinguish the feminist emancipatory core from other attempts of developing and improving society in various ways. In this workshop, we therefore want to revisit our feminist roots, where theory is a liberatory and creative practice, motivated by affect, curiosity, and wonder. From this standpoint, we consider which of our feminist tools can make a significant difference today, in a highly datafied world. The goal of this workshop is to;1) create an inventory of feminist theories and concepts that have had an impact on our work as designers, educators, researchers, and activists;2) develop a feminist toolbox for the CSCW community to strengthen our feminist literacy. © 2023 Owner/Author.

关键词： Computation theory