Truth Discovery and Crowdsourcing

This line of work deals with statistical methods to aggregate information from many sources in an attempt to discover the underlying ground truth, focusing on complex labels (beyond multiple-choice and real-valued).

  • Truth Discovery via Proxy Voting. Reshef Meir, Ofra Amir, Gal Cohensius, Omer Ben-Porat, and Lirong Xia. AAAI’23 [Arxiv← start here
  • Efficient Online Crowdsourcing with Complex Annotations. Reshef Meir, Viet-An Nguyen, Xu Chen, Jagdish Ramakrishnan and Udi Weinsberg. AAAI 2024. [Arxiv]
  • Empirical Bayes approach to Truth Discovery problems. Tsviel Ben-Shabat, Reshef Meir, and David Azriel, UAI ’22. [Arxiv]
  • When is Proximity-based Truth Discovery Good Enough? Doron Kabla and Reshef Meir[Working paper]
  • Maximum Likelihood Voting Rules for Heterogeneous Voters. Doron Kabla and Reshef Meir. [Working paper]