Yevgeni Berzak

4 paper cards in Mova's library · 2016–2022

Yevgeni Berzak wrote 4 of the papers Mova reads, in 2016–2022, most of them on Dependency parsing, Universal Dependencies, Treebanks; often with Boris Katz, Adam Yaari, Andrei Barbu.

The profile is the library's match of the name; two people with one name are not always told apart.

Ideas Mova took (1)

From the ideas ledger: an idea tried, kept or queued, or measured with no gain, and what it gave.

kept — Critical-span ablation as a validity check of course questions: remove the critical sentence; the student should fall to guessing on that question. Questions still answered right are guessable from elsewhere or from the options and should be fixed or marked. (STARC: Structured Annotations for Reading Comprehension · critical-span ablation per lesson as the «key sentences removed» control (dba7f8c6e))

The Aligned Multimodal Movie Treebank: An audio, video, dependency-parse treebank (2022)
The Aligned Multimodal Movie Treebank (AMMT) takes dialogue from Hollywood films, with word-level alignment to the audio and video, POS tags and UD dependency parses.
STARC: Structured Annotations for Reading Comprehension (2020) 1 idea kept
STARC is a scheme for four-option reading questions in which the options have fixed roles tied to marked text spans: A is right and rests on a «critical span»; B mentions the critical span but misunderstands it; C refers to a separate «distractor span»; D has no support in the text.
Classifying Syntactic Errors in Learner Language (2020)
The authors present SERCL, an automatic taxonomy and classifier for syntactic errors in learner language, where a syntactic error is one whose correction changes the part of speech, dependency relation or morphological features involved.
Universal Dependencies for Learner English (2016)
Much of the English written today is produced by non-native speakers, yet before this work there was no open syntactic treebank of learner English.