AICLAug 5, 2016

Winograd Schemas and Machine Translation

arXiv:1608.01884v27 citations
Originality Synthesis-oriented
AI Analysis

This addresses the problem of evaluating and improving machine translation systems for pronoun disambiguation, but it appears incremental as it applies existing Winograd schemas to a new context.

The paper explores using Winograd schemas, which involve ambiguous pronouns requiring commonsense knowledge, as challenges for machine translation by leveraging pronoun distinctions like gender in target languages not present in the source.

A Winograd schema is a pair of sentences that differ in a single word and that contain an ambiguous pronoun whose referent is different in the two sentences and requires the use of commonsense knowledge or world knowledge to disambiguate. This paper discusses how Winograd schemas and other sentence pairs could be used as challenges for machine translation using distinctions between pronouns, such as gender, that appear in the target language but not in the source.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes