Scim: Intelligent Faceted Highlights for Interactive, Multi-Pass Skimming of Scientific Papers

Researchers are expected to keep up with an immense literature, yet often find it prohibitively time-consuming to do so. This paper ex-plores how intelligent agents can help scaffold in-situ information seeking across scientific papers. Specifically, we present Scim, an AI-augmented reading interface designed to help researchers skim papers by automatically identifying, classifying, and highlighting salient sentences, organized into rhetorical facets rooted in common information needs. Using Scim as a design probe, we explore the benefits and drawbacks of imperfect AI assistance within an augmented reading interface. We found researchers used Scim in several different ways: from reading primarily in the ‘highlight browser’ (side panel) to making multiple passes through the paper with different facets activated (e.g., focusing solely on objective and novelty in their first pass). From our study, we identify six key design recommendations and avenues for future research in augmented reading interfaces.

[1]  C. Lee Giles,et al.  SciBERTSUM: Extractive Summarization for Scientific Documents , 2022, DAS.

[2]  Vibhav Gogate,et al.  Anchoring Bias Affects Mental Model Formation and User Reliance in Explainable AI Systems , 2021, IUI.

[3]  Marti A. Hearst,et al.  Augmenting Scientific Papers with Just-in-Time, Position-Sensitive Definitions of Terms and Symbols , 2020, CHI.

[4]  Robert Y. Wang,et al.  Decoding Surface Touch Typing from Hand-Tracking , 2020, UIST.

[5]  Shay B. Cohen,et al.  Reducing the Frequency of Hallucinated Quantities in Abstractive Summaries , 2020, FINDINGS.

[6]  Daniel S. Weld,et al.  TLDR: Extreme Summarization of Scientific Documents , 2020, FINDINGS.

[7]  Edward Lank,et al.  Chameleon: Bringing Interactivity to Static Digital Documents , 2020, CHI.

[8]  Feng Xia,et al.  Scientific Paper Recommendation: A Survey , 2020, IEEE Access.

[9]  Giuseppe Carenini,et al.  Extractive Summarization of Long Documents by Combining Global and Local Context , 2019, EMNLP.

[10]  Bhavana Dalvi,et al.  Pretrained Language Models for Sequential Sentence Classification , 2019, EMNLP.

[11]  Jeffrey P. Bigham,et al.  StateLens: A Reverse Engineering Solution for Making Existing Dynamic Touchscreens Accessible , 2019, UIST.

[12]  Paul N. Bennett,et al.  Guidelines for Human-AI Interaction , 2019, CHI.

[13]  Walid-Khaled Hidouci,et al.  Automatic text summarization: What has been done and what has to be done , 2019, ArXiv.

[14]  Niklas Elmqvist,et al.  Elastic Documents: Coupling Text and Tables through Contextual Visualizations for Enhanced Document Reading , 2019, IEEE Transactions on Visualization and Computer Graphics.

[15]  Yuanchun Shi,et al.  Lip-Interact: Improving Mobile Device Interaction with Silent Speech Commands , 2018, UIST.

[16]  Maneesh Agrawala,et al.  Facilitating Document Reading by Linking Text and Tables , 2018, UIST.

[17]  Albrecht Schmidt,et al.  Skim-reading Strategies in Sighted and Visually-Impaired Individuals: A Comparative Study , 2018, PETRA.

[18]  Doug Downey,et al.  Construction of the Literature Graph in Semantic Scholar , 2018, NAACL.

[19]  Nan Hua,et al.  Universal Sentence Encoder , 2018, ArXiv.

[20]  Sen Wang,et al.  HiText: Text Reading with Dynamic Salience Marking , 2017, WWW.

[21]  David R. Karger,et al.  Wikum: Bridging Discussion Forums and Wikis Using Recursive Summarization , 2017, CSCW.

[22]  Antti Oulasvirta,et al.  Spotlights: Attention-Optimized Highlights for Skim Reading , 2016, CHI.

[23]  Rebecca Treiman,et al.  So Much to Read, So Little Time , 2016, Psychological science in the public interest : a journal of the American Psychological Society.

[24]  Bela Gipp,et al.  Research-paper recommender systems: a literature survey , 2015, International Journal on Digital Libraries.

[25]  Liangcai Gao,et al.  Scientific Information Understanding via Open Educational Resources (OER) , 2015, SIGIR.

[26]  Rubaiat Habib Kazi,et al.  Your Paper is Dead!: Bringing Life to Research Articles with Animated Figures , 2015, CHI Extended Abstracts.

[27]  Krzysztof Z. Gajos,et al.  Content-aware kinetic scrolling for supporting web page navigation , 2014, UIST.

[28]  T. Kuflik,et al.  Semantize: visualizing the sentiment of individual document , 2014, AVI.

[29]  Maneesh Agrawala,et al.  Extracting references between text and charts via crowdsourcing , 2014, CHI.

[30]  Ji Soo Yi.  QnDReview: read 100 CHI papers in 7 hours , 2014, CHI Extended Abstracts.

[31]  Anna Korhonen,et al.  Improved Information Structure Analysis of Scientific Documents Through Discourse and Lexical Constraints , 2013, NAACL.

[32]  J. B. Brooke,et al.  SUS: a retrospective , 2013 .

[33]  Dietrich Rebholz-Schuhmann,et al.  Automatic recognition of conceptualization zones in scientific articles and two life science applications , 2012, Bioinform..

[34]  Ani Nenkova,et al.  A Survey of Text Summarization Techniques , 2012, Mining Text Data.

[35]  Stephen J. Payne,et al.  Skim reading by satisficing: evidence from eye tracking , 2011, CHI.

[36]  W. Keith Edwards,et al.  Active reading and its discontents: the situations, problems and ideas of readers , 2011, CHI.

[37]  Simone Teufel,et al.  Corpora for the Conceptualisation and Zoning of Scientific Papers , 2010, LREC.

[38]  Thanos Athanasiou,et al.  How to Read a Paper , 2010 .

[39]  George Buchanan,et al.  An Empirical Study of User Navigation during Document Triage , 2009, ECDL.

[40]  Geoffrey B. Duggan,et al.  Text skimming: the process and effectiveness of foraging through text under time pressure. , 2009, Journal of experimental psychology. Applied.

[41]  Simone Teufel,et al.  Towards Domain-Independent Argumentative Zoning: Evidence from Chemistry and Computational Linguistics , 2009, EMNLP.

[42]  Philip T. Kortum,et al.  Determining what individual SUS scores mean: adding an adjective rating scale , 2009 .

[43]  Robert Dale,et al.  Enriching a document collection by integrating information extraction and PDF annotation , 2009, Electronic Imaging.

[44]  Jöran Beel,et al.  Google Scholar’s Ranking Algorithm : An Introductory Overview , 2009 .

[45]  Carol Tenopir,et al.  Electronic Journals and Changes in Scholarly Article Seeking and Reading Patterns , 2008, D Lib Mag..

[46]  George Buchanan,et al.  Improving skim reading for document triage , 2008, IIiX.

[47]  Naoaki Okazaki,et al.  Identifying Sections in Scientific Abstracts using Conditional Random Fields , 2008, IJCNLP.

[48]  Stephen J. Payne,et al.  Allocating Time Across Multiple Texts: Sampling and Satisficing , 2007, Hum. Comput. Interact..

[49]  Ed H. Chi,et al.  Visual Foraging of Highlighted Text: An Eye-Tracking Study , 2007, HCI.

[50]  Paul Dourish,et al.  How HCI interprets the probes , 2007, CHI.

[51]  Steven K. Feiner,et al.  Content-aware scrolling , 2006, UIST.

[52]  Ziming Liu,et al.  Reading behavior in the digital environment: Changes in reading behavior over the past ten years , 2005, J. Documentation.

[53]  Ed H. Chi,et al.  ScentHighlights: highlighting conceptually-related sentences during reading , 2005, IUI.

[54]  Regina A. Pomranky,et al.  The role of trust in automation reliance , 2003, Int. J. Hum. Comput. Stud..

[55]  Marc Moens,et al.  Articles Summarizing Scientific Articles: Experiments with Relevance and Rhetorical Status , 2002, CL.

[56]  Natasha Lacroix.  Macrostructure construction and organization in the processing of multiple text passages , 1999 .

[57]  Eric Horvitz,et al.  Principles of mixed-initiative user interfaces , 1999, CHI '99.

[58]  Jamey Graham,et al.  The reader's helper: a personalized document reading environment , 1999, CHI '99.

[59]  Takeo Igarashi,et al.  A negotiation architecture for fluid documents , 1998, UIST '98.

[60]  Jock D. Mackinlay,et al.  Fluid links for informed and incremental link transitions , 1998, HYPERTEXT '98.

[61]  Neville Moray,et al.  Trust and human intervention in automated systems , 1995 .

[62]  Bonnie M. Muir,et al.  Trust in automation. I: Theoretical issues in the study of trust and human intervention in automated systems , 1994 .

[63]  Donald A. Norman,et al.  How might people interact with agents , 1994, CACM.

[64]  James D. Hollan,et al.  Edit wear and read wear , 1992, CHI.

[65]  Patricia L. Carrell,et al.  Facilitating ESL Reading by Teaching Text Structure , 1985 .

[66]  M. Masson.  Conceptual processing of text during skimming and rapid sequential reading , 1983, Memory & cognition.

[67]  Michael E. J. Masson,et al.  Cognitive Processes in Skimming Stories , 1982 .

[68]  John P. Rickards.  Notetaking, Underlining, Inserted Questions, and Organizers in Text: Research Conclusions and Educational Implications. , 1980 .

[69]  R. Fowler,et al.  Effectiveness of highlighting for retention of text material. , 1974 .

[70]  M. Maxwell.  Skimming and Scanning Improvement: The Needs, Assumptions and Knowledge Base , 1972 .

[71]  H. V. Restorff.  Über die Wirkung von Bereichsbildungen im Spurenfeld , 1933 .