https://hackerpaper.kaipfister.de/p/1606.02270/Collected HN comments and submissions linking this paper.enMon, 02 Nov 2026 11:16:22 GMThttps://news.ycombinator.com/item?id=11865075urn:hackerpapers:mention:1606.02270:11865075Wed, 08 Jun 2016 19:49:05 GMT<p>You are correct - both teams are Canadian and part of Maluuba Research Lab in Montreal. [www.maluuba.com] Here is the difference between the two: The EpiReader model uses a two-stage process to arrive at an answer. The first stage extracts a small set of candidate answers from the text passage and the second stage turns these into hypotheses and then tests them through neural inference. [http://arxiv.org/abs/1606.02270] The Alternating Iterative Attention model goes through a dynamic procedure of focusing on different parts of the question and the document in turn.…</p><p><a href="https://hackerpaper.kaipfister.de/p/1606.02270/">Paper and discussion history</a>. A mention is not an endorsement.</p>cs.CLhttps://news.ycombinator.com/item?id=11863995urn:hackerpapers:mention:1606.02270:11863995Wed, 08 Jun 2016 17:37:37 GMT<p>Here is the paper: http://arxiv.org/abs/1606.02270 &quot;EpiReader does this using two neural networks, a type of AI inspired by how neurons work in the human brain. The first neural net picks a set of likely answers based on its understanding of the paragraph. The second evaluates the reasoning used by the first to come up with the right answer.&quot;</p><p><a href="https://hackerpaper.kaipfister.de/p/1606.02270/">Paper and discussion history</a>. A mention is not an endorsement.</p>cs.CL