Sharing personal informatics data can support accountability, connectedness, and self-expression, but people often find their data too trivial to share on social media. Ephemeral social platforms like Snapchat and Story features have emerged as spaces for sharing more trivial life events, presenting an opportunity to incorporate self-tracked data into sharing. Past work suggests that including data-driven stickers on these platforms can help add additional context to what people share, but little is known about the benefits and challenges of people's everyday experiences with this concept. To understand people's everyday use of data-driven stickers, we designed and developed SnapPI, an app for flexibly incorporating data into stickers for Snapchat. We deployed SnapPI to 21 participants for two weeks, finding that participants value aligning data sharing with Snapchat's communication and stylistic norms. Perceiving Snapchat as a playful platform, participants connected data stickers with various visual components of their Snaps. Stickers were used to incorporate personal informatics data into their existing conversations, and were edited to align with different audience needs or to be expressive. We discuss recommendations for personal data sharing, suggesting supporting flexibility in presentation and aligning with the norms of existing platforms.

SnapPI: Understanding Everyday Use of Personal Informatics Data Stickers on Ephemeral Social Media

Tracking progress in machine learning has become increasingly difficult with the recent explosion in the number of papers. In this paper, we present AxCell, an automatic machine learning pipeline for extracting results from papers. AxCell uses several novel components, including a table segmentation subtask, to learn relevant structural knowledge that aids extraction. When compared with existing methods, our approach significantly improves the state of the art for results extraction. We also release a structured, annotated dataset for training models for results extraction, and a dataset for evaluating the performance of models on this task. Lastly, we show the viability of our approach enables it to be used for semi-automated results extraction in production, suggesting our improvements make this task practically viable for the first time. Code is available on GitHub.

AxCell: Automatic Extraction of Results from Machine Learning Papers

/pdf/axcell-automatic-extraction-of-results-from-machine-learning-4w0bknqhh8.pdf

AxCell: Automatic Extraction of Results from Machine Learning Papers.

Over the years, researchers capitalized on the repetitiveness of software changes to automate many software evolution tasks. Despite the extraordinary rise in popularity of Python-based ML systems, they do not benefit from these advances. Without knowing what are the repetitive changes that ML developers make, researchers, tool, and library designers miss opportunities for automation, and ML developers fail to learn and use best coding practices. To fill the knowledge gap and advance the science and tooling in ML software evolution, we conducted the first and most fine-grained study on code change patterns in a diverse corpus of 1000 top-rated ML systems comprising 58 million SLOC. To conduct this study we reuse, adapt, and improve upon the state-of-the-art repetitive change mining techniques. Our novel tool, R-CPATMINER, mines over 4M commits and constructs 350K fine-grained change graphs and detects 28K change patterns. Using thematic analysis, we identified 22 pattern groups and we reveal 4 major trends of how ML developers change their code. We surveyed 650 ML developers to further shed light on these patterns and their applications, and we received a 15% response rate. We present actionable, empirically-justified implications for four audiences: (i) researchers, (ii) tool builders, (iii) ML library vendors, and (iv) developers and educators.

/pdf/discovering-repetitive-code-changes-in-python-ml-systems-2ow215t3.pdf

Discovering Repetitive Code Changes in Python ML Systems

In modern recommender systems, there are usually comments or reviews from users that justify their ratings for different items. Trained on such textual corpus, explainable recommendation models learn to discover user interests and generate personalized explanations. Though able to provide plausible explanations, existing models tend to generate repeated sentences for different items or empty sentences with insufficient details. This begs an interesting question: can we immerse the models in a multimodal environment to gain proper awareness of real-world concepts and alleviate above shortcomings? To this end, we propose a visually-enhanced approach named METER with the help of visualization generation and text–image matching discrimination: the explainable recommendation model is encouraged to visualize what it refers to while incurring a penalty if the visualization is incongruent with the textual explanation. Experimental results and a manual assessment demonstrate that our approach can improve not only the text quality but also the diversity and explainability of the generated explanations.

/pdf/improving-personalized-explanation-generation-through-1b81auzp.pdf

Improving Personalized Explanation Generation through Visualization

We introduce an AI-enabled portal that presents an excellent visualization of Mahatma Gandhi's life events by constructing temporal and spatial social networks from the Gandhian literature. Applying an ensemble of methods drawn from NLTK, Polyglot and Spacy we extract the key persons and places that find mentions in Gandhi's written works. We visualize these entities and connections between them based on co-mentions within the same time frame as networks in an interactive web portal. The nodes in the network, when clicked, fire search queries about the entity and all the information about the entity presented in the corresponding book from which the network is constructed, are retrieved and presented back on the portal. Overall, this system can be used as a digital and user-friendly resource to study Gandhian literature.

Gandhipedia: A one-stop AI-enabled portal for browsing Gandhian literature, life-events and his social network

A leaderboard is a tabular presentation of performance scores of the best competing techniques that address a specific scientific problem. Manually maintained leaderboards take time to emerge, which induces a latency in performance discovery and meaningful comparison. This can delay dissemination of best practices to non-experts and practitioners. Regarding papers as proxies for techniques, we present a new system to automatically discover and maintain leaderboards in the form of partial orders between papers, based on performance reported therein. In principle, a leaderboard depends on the task, data set, other experimental settings, and the choice of performance metrics. Often there are also tradeoffs between different metrics. Thus, leaderboard discovery is not just a matter of accurately extracting performance numbers and comparing them. In fact, the levels of noise and uncertainty around performance comparisons are so large that reliable traditional extraction is infeasible. We mitigate these challenges by using relatively cleaner, structured parts of the papers, e.g., performance tables. We propose a novel performance improvement graph with papers as nodes, where edges encode noisy performance comparison information extracted from tables. Every individual performance edge is extracted from a table with citations to other papers. These extractions resemble (noisy) outcomes of ‘matches’ in an incomplete tournament. We propose several approaches to rank papers from these noisy ‘match’ outcomes. We show that our ranking scheme can reproduce various manually curated leaderboards very well. Using widely-used lists of state-of-the-art papers in 27 areas of Computer Science, we demonstrate that our system produces very reliable rankings. We also show that commercial scholarly search systems cannot be used for leaderboard discovery, because of their emphasis on citations, which favors classic papers over recent performance breakthroughs. Our code and data sets will be placed in the public domain.

Automated early leaderboard generation from comparative tables

A leaderboard is a tabular presentation of performance scores of the best competing techniques that address a specific scientific problem. Manually maintained leaderboards take time to emerge, which induces a latency in performance discovery and meaningful comparison. This can delay dissemination of best practices to non-experts and practitioners. Regarding papers as proxies for techniques, we present a new system to automatically discover and maintain leaderboards in the form of partial orders between papers, based on performance reported therein. In principle, a leaderboard depends on the task, data set, other experimental settings, and the choice of performance metrics. Often there are also tradeoffs between different metrics. Thus, leaderboard discovery is not just a matter of accurately extracting performance numbers and comparing them. In fact, the levels of noise and uncertainty around performance comparisons are so large that reliable traditional extraction is infeasible. We mitigate these challenges by using relatively cleaner, structured parts of the papers, e.g., performance tables. We propose a novel performance improvement graph with papers as nodes, where edges encode noisy performance comparison information extracted from tables. Every individual performance edge is extracted from a table with citations to other papers. These extractions resemble (noisy) outcomes of 'matches' in an incomplete tournament. We propose several approaches to rank papers from these noisy 'match' outcomes. We show that our ranking scheme can reproduce various manually curated leaderboards very well. Using widely-used lists of state-of-the-art papers in 27 areas of Computer Science, we demonstrate that our system produces very reliable rankings.

Atharva Vyas

Papers

Gandhipedia: A one-stop AI-enabled portal for browsing Gandhian literature, life-events and his social network

Automated early leaderboard generation from comparative tables

Automated Early Leaderboard Generation From Comparative Tables

Gandhipedia: A One-stop AI-enabled Portal for Browsing Gandhian Literature, Life-events and His Social Network