Publications

TLDRCocoa is presented, a system that introduces a novel design pattern -- interactive plans -- for collaborating with an AI agent on complex, multi-step tasks and saw the interleaving of co-planning and co-execution as an effective novel paradigm for human-AI collaboration. Expand

TLDROpenScholar is introduced, a specialized retrieval-augmented language model that answers scientific queries by identifying relevant passages from 45 million open-access papers and synthesizing citation-backed responses and improves off-the-shelf LMs by 12%.

TLDRThe first large-scale survey of 816 verified research article authors is presented, finding that traditionally disadvantaged groups in academia (non-White, junior, and non-native English speaking researchers) report higher LLM usage and perceived benefits, suggesting potential for improved research equity. Expand

TLDRThis work presents an approach for augmenting unannotated table corpora withSynthesized intents, and applies it to create a dataset for studying schema generation conditioned on a given information need, thus reducing ambiguity and comprehensively benchmarking several single-shot schema generation methods. Expand

AbstractThe continuous, rapid development of general-purpose models like LLMs suggests the theoretical possibility of AI performing any human task. Yet, despite the potential and promise, these models are far from perfect, excelling at certain tasks while st... Expand

TLDRIt is demonstrated that participants effectively used IdeaSynth for real-world research projects at various ideation stages from developing initial ideas to revising framings of mature manuscripts, highlighting the possibilities to adopt IdeaSynth in researcher’s workflows. Expand

TLDRDimInd is an interactive system that scaffolds literature review across large paper collections through LLM-generated structured representations through LLM-generated structured representations and supported participants in extracting information and conceptually organizing papers with less effort compared to a ChatGPT-assisted baseline workflow. Expand

TLDRThis work introduces Ai2 Scholar QA, a free online scientific question answering application that outperforms competing systems on a recent scientific QA benchmark and makes the entire pipeline public to facilitate research. Expand

TLDRSciArena is presented, an open and collaborative platform for evaluating foundation models on scientific literature-grounded tasks, and SciArena-Eval, a meta-evaluation benchmark based on collected preference data, which measures the accuracy of models in judging answer quality by comparing their pairwise assessments with human votes. Expand

TLDRThis work presents Social-RAG, a workflow for socially grounding agents that retrieves context from prior group interactions, selects relevant social signals, and feeds them into a language model to generate messages in a socially aligned manner. Expand

TLDRA framework that leverages LMs to perform this task by decomposing it into separate schema and value generation steps is introduced, and it is found that even when LMs fail to fully reconstruct a reference table, their generated novel aspects can still be useful. Expand

TLDRThis work explores a design space of meronymous communication, where people can reveal carefully chosen aspects of their identity and also leverage trusted endorsers to gain credibility, in a system for scholars to meronymously seek and receive paper recommendations on Twitter and Mastodon. Expand

TLDRIt was shown that participants using PaperWeaver were able to better understand the relevance of recommended papers and triage them more confidently when compared to a baseline that presented the related work sections from recommended papers. Expand

TLDROpenScholar is introduced, a specialized retrieval-augmented LM that answers scientific queries by identifying relevant passages from 45 million open-access papers and synthesizing citation-backed responses, and achieves citation accuracy on par with human experts. Expand

Abstract Language model users often issue queries that lack specification, where the context under which a query was issued—such as the user’s identity, the query’s intent, and the criteria for a response to be useful—is not explicit. For instance, a good r... Expand

TLDRThis work proposes a design space as a structured way to examine and explore the multidimensional space of intelligent and interactive writing assistants, and explores five aspects of writing assistants: task, user, technology, interaction, and ecosystem. Expand

TLDRA novel computational pipeline is developed that ties together user input of relevant seed threads with citation graphs and LLMs, to expand and structure them, respectively, and is found to help scholars efficiently make sense of relevant threads, broaden their perspectives, and increases their curiosity. Expand

TLDRRecursively expandable abstracts are introduced, a novel interaction paradigm that dynamically expands abstracts by progressively incorporating additional information from the papers’ full text during their dive into the full text. Expand

TLDRThe Semantic Reader Project is described, a collaborative effort across multiple institutions to explore automatic creation of dynamic reading interfaces for research papers, and a collection of novel reading interfaces are developed and evaluated and evaluated them with study participants and real-world users to show improved reading experiences for scholars. Expand

AbstractDespite growing interest in applying natural language processing (NLP) and computer vision (CV) models to the scholarly domain, scientific documents remain challenging to work with. They’re often in difficult-to-use PDF formats, and the ecosystem of ... Expand

TLDRPapeos is presented, a novel reading and authoring interface that allow authors to augment their papers by segmenting and localizing talk videos alongside relevant paper passages with automatically generated suggestions. Expand

TLDRCiteSee is a paper reading tool that leverages a user’s publishing, reading, and saving activities to provide personalized visual augmentations and context around citations to help users prioritize their exploration. Expand

TLDRIt is demonstrated how ComLittee improves author and paper discovery, a literature discovery system that supports author-centric exploration in contrast to paper-centric interaction in prior systems. Expand

TLDRThis work designs a system, Relatedly, that scaffolds exploring and reading multiple related work paragraphs on a topic, with features including dynamic re-ranking and highlighting to spotlight unexplored dissimilar information, auto-generated descriptive paragraph headings, and low-lighting of redundant information. Expand

TLDRThis research investigates features representing domain, subdomain, and individual knowledge to predict individual jargon familiarity and compares supervised and prompt-based approaches, finding that prompt-based methods yield the highest accuracy, though the task remains difficult and supervised approaches have lower false positive rates. Expand

TLDRIt is argued that developing AI supports for expository writing has unique and exciting research challenges and can lead to high real-world impacts. Expand

TLDRA tool integrated into users’ reading process that helps them with leveraging authors’ existing summarization of threads, typically in introduction or related work sections, in order to situate their own work’s contributions is developed. Expand

TLDRFuse is introduced, a browser extension that externalizes users’ working memory by combining low-cost collection with lightweight organization of content in a compact card-based sidebar that is always available and discusses how these affordances help users externalize more of their mental model into the system. Expand

TLDRA new interaction technique called wiggling is explored, which can be used to fluidly collect, organize, and rate information during early sensemaking stages with a single gesture, with a 58% reduction in operational cost while being 24% faster compared to a common baseline. Expand

TLDRA Chrome extension called Tabs.do is introduced, which explores bringing a task-centric approach to the browser, helping users to group their tabs into tasks and then organize, prioritize, and switch between those tasks fluidly. Expand

TLDRHow tabs today are overloaded with a diverse set of functionalities and issues users face when managing them is investigated and design implications for future browser interfaces that can better support managing these pressures are developed. Expand

TLDRMesh bridges the gap between decision support systems that typically have rigid structures and the fluid and dynamic process of exploratory search, changing the cost structure to provide increasing payoffs with greater user investment. Expand

TLDRResults from a controlled lab study suggest that the approach incentivized participants to express their interests more richly than in a baseline condition, and a field study showed that participants found benefits in SearchLens while conducting their own tasks. Expand

TLDRSOLVENT is introduced, a mixed-initiative system where humans annotate aspects of research papers that denote their background, purpose, mechanism, and findings, and a computational model constructs a semantic representation from these annotations that can be used to find analogies among the research papers. Expand

TLDREvolent, a crowd-powered conversational assistant built to automate itself over time by allowing new chatbots to be easily integrated to automate more scenarios, and reusing prior crowd answers, and learning to automatically approve response candidates is introduced. Expand

TLDRThis work introduces a new way of browsing through a scaffolded interface in the Bento mobile search system, finding converging evidence that users were able to make progress on their complex searching tasks with this structure, and find it more organized and easier to revisit.

TLDRRevolt eliminates the burden of creating detailed label guidelines by harnessing crowd disagreements to identify ambiguous concepts and create rich structures (groups of semantically related items) for post-hoc label decisions. Expand

TLDRThis work introduces the idea of intentionally supporting uncertain input in the context of saving information during complex reading and information exploration in a system that uses force touch and fuzzy bounding boxes along with posthoc expandable context to support identifying and saving information in an intentionally uncertain way on mobile devices.

TLDRThis paper instantiates the idea that a computational system can scaffold an emerging interdependent, big picture view entirely through the small contributions of individuals through a prototype system for accomplishing distributed information synthesis and evaluates its output across a variety of topics.

TLDRAlloy, a hybrid approach that combines the richness of human judgments with the power of machine algorithms, is introduced, a modular "cast and gather" approach which leverages a machine learning backbone to stitch together different types of judgment tasks.