• Publications
  • Influence
SOLVENT: A Mixed Initiative System for Finding Analogies between Research Papers
TLDR
SOLVENT 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
SciSight: Combining faceted navigation and research group detection for COVID-19 exploratory scientific search
TLDR
Sight is presented, a system for exploratory search of COVID-19 research integrating two key capabilities: first, exploring associations between biomedical facets automatically extracted from papers; second, combining textual and network information to search and visualize groups of researchers and their ties. Expand
Scaling up analogical innovation with crowds and AI
TLDR
A perspective on the future of scalable analogical innovation and first steps using crowds and artificial intelligence to augment creativity that quantitatively demonstrate the promise of the approach, as well as core challenges critical to realizing this vision are laid out. Expand
Accelerating Innovation Through Analogy Mining
TLDR
This approach combines crowdsourcing and recurrent neural networks to extract purpose and mechanism vector representations from product descriptions and demonstrates that these learned vectors allow us to find analogies with higher precision and recall than traditional methods. Expand
SciSight: Combining faceted navigation and research group detection for COVID-19 exploratory scientific search
TLDR
Sight is presented, a system for exploratory search of COVID-19 research integrating two key capabilities: first, exploring associations between biomedical facets automatically extracted from papers; second, combining textual and network information to search and visualize groups of researchers and their ties. Expand
Accelerating Innovation Through Analogy Mining
TLDR
This paper explores the viability and value of learning simpler structural representations, specifically, "problem schemas", which specify the purpose of a product and the mechanisms by which it achieves that purpose, and combines crowdsourcing and recurrent neural networks to extract purpose and mechanism vector representations from product descriptions. Expand
SOLVENT
Scientific discoveries are often driven by finding analogies in distant domains, but the growing number of papers makes it difficult to find relevant ideas in a single discipline, let alone distantExpand
Ballpark Learning: Estimating Labels from Rough Group Comparisons
TLDR
Across several domains, it is shown how using only proportion constraints and no labeled examples, this work can achieve surprisingly high accuracy in estimating individual labels given only coarse, aggregated signal over the data points. Expand
Extracting a Knowledge Base of Mechanisms from COVID-19 Papers
TLDR
This work pursues the construction of a knowledge base of mechanisms—a fundamental concept across the sciences, which encompasses activities, functions and causal relations, ranging from cellular processes to economic impacts, by developing a broad, unified schema. Expand
Language (Re)modelling: Towards Embodied Language Understanding
TLDR
It is argued that the use of grounding by metaphoric reasoning and simulation will greatly benefit NLU systems, and a system architecture along with a roadmap towards realizing this vision is proposed. Expand
...
1
2
3
...