• UMass IESL
  • News
  • People
    • Director
    • Postdocs
    • PhD Students
    • Masters Students
    • Undergraduates
    • Staff
    • Alumni: PhD
    • Alumni: Postdoc
    • Alumni: Staff
    • Alumni: MS & UG
  • Research
  • Projects
  • Publications
  • Software
  • Datasets
    • UMass Citation Field Extraction Dataset
    • Wikilinks Dataset
    • BibTex Dataset
  • Social
  • Wiki (internal)

News

  • LISA has won the best paper award at EMNLP 2018!. Congrats Emma and Pat!
  • Linguistically-Informed Self-Attention for Semantic Role Labeling (LISA) and Marginal Likelihood Training of BiLSTM-CRF for Biomedical Named Entity Recognition from Disjoint Label Sets are accepted to EMNLP 2018.
  • Compact Representation of Uncertainty In Clustering has been accepted to NIPS 2018.
  • Embedded-State Latent Conditional Random Fields for Sequence Labeling is accepted to CoNLL 2018.
  • Andrew McCallum appointed as a distinguished professor. Congrats Andrew!
  • Box embeddings and hierarchical losses and new resource for fine grained entity typing and linking is accepted to ACL 2018!
  • 3 papers accepted to NAACL 2018! - Bio Medical Relation Extraction, Unsupervised Hypernymy Detection and training SPEN’s with indirect supervision.
  • MINERVA is accepted to ICLR 2018.
  • Andrew McCallum will be leading the Computable Knowledge project, a new partnership with Chan Zuckerberg Initiative.
  • Andrew McCallum has been named as an 2017 ACM fellow. Congrats Andrew!
  • Emma Strubell receives the IBM Ph.D. Fellowship award. Congrats Emma!
  • MINERVA, a new way of reasoning on large knowledge bases using reinforcement learning has won the best paper award at AKBC at NIPS 2017.
  • Active Bias: Training more accurate neural networks by emphasizing high variance samples has been accepted to NIPS 2017.
  • Iterated Dilated Convolutions, a faster and better alternative to Bi-LSTMs for named entity recognition has been accepted to EMNLP 2017.
  • PERCH, a new non-greedy algorithm for online hierarchical clustering that scales to both massive number of samples and number of clusters has accepted to KDD 2017.
  • QA on Knowledge Bases and Text using Universal Schema and Memory Networks has been accepted to ACL 2017.
  • Structured Prediction Energy Networks [Belanger, McCallum ICML 2016] are an alternative to graphical models, leveraging deep learning to discover rich dependencies among output variables.
  • Our research on universal schema is currently at the top of the Stanford KBP leaderboard! Congratulations to Haw-Shiuan Chang, Pat Verga, Emma Strubell, Nick Monath, and the other IESL students who worked on this.
  • Arvind Neelakantan receives the Google Ph.D. fellowship in Machine Learning.
  • OpenReview.net is hosting reviewing for ICLR 2018. It has previously hosted ICLR 2017, UAI 2017.
  • FACTORIE is a toolkit for deployable probabilistic modeling, implemented as a software library in Scala. It provides its users with a succinct language for creating relational factor graphs, estimating parameters and performing inference.
Follow @andrewmccallum Tweets by andrewmccallum Follow @UMassCS Tweets by UMassCICS

© UMass IESL 2018

College of Information and Computer Sciences, University of Massachusetts Amherst.