Share This Author
Deterministic Coreference Resolution Based on Entity-Centric, Precision-Ranked Rules
- Heeyoung Lee, Angel X. Chang, Yves Peirsman, Nathanael Chambers, M. Surdeanu, Dan Jurafsky
- Computer ScienceCL
- 1 December 2013
The two stages of the sieve-based architecture, a mention detection stage that heavily favors recall, followed by coreference sieves that are precision-oriented, offer a powerful way to achieve both high precision and high recall.
Stanford’s Multi-Pass Sieve Coreference Resolution System at the CoNLL-2011 Shared Task
- Heeyoung Lee, Yves Peirsman, Angel X. Chang, Nathanael Chambers, M. Surdeanu, Dan Jurafsky
- Computer ScienceCoNLL Shared Task
- 23 June 2011
The coreference resolution system submitted by Stanford at the CoNLL-2011 shared task was ranked first in both tracks, with a score of 57.8 in the closed track and 58.3 in the open track.
A Multi-Pass Sieve for Coreference Resolution
This work proposes a simple coreference architecture based on a sieve that applies tiers of deterministic coreference models one at a time from highest to lowest precision, and outperforms many state-of-the-art supervised and unsupervised models on several standard corpora.
Joint Entity and Event Coreference Resolution across Documents
- Heeyoung Lee, Marta Recasens, Angel X. Chang, M. Surdeanu, Dan Jurafsky
- Computer ScienceEMNLP
- 12 July 2012
A novel coreference resolution system that models entities and events jointly that handles nominal and verbal events as well as entities, and the joint formulation allows information from event coreference to help entity coreference, and vice versa.
On the Importance of Text Analysis for Stock Price Prediction
A system that forecasts companies stock price changes (UP, DOWN, STAY) in response to financial events reported in 8-K documents is introduced, indicating that using text boosts prediction accuracy over 10% over a strong baseline that incorporates many financially-rooted features.
A scaffolding approach to coreference resolution integrating statistical and rule-based models
A scaffolding approach to the task of coreference resolution that incrementally combines statistical classifiers, each designed for a particular mention type, with rule-based models (for sub-tasks well-matched to determinism) and achieves a runtime speedup of 550 per cent without considerable loss of accuracy.
Using Out-of-Domain Data for Lexical Addressee Detection in Human-Human-Computer Dialog
This study investigates how well out-of-domain data (either outside the domain, or from single-user scenarios) can fill in for matched in- domain data and finds that human-addressed speech can be modeled using out- of-domain conversational speech transcripts, and thathuman-computer utterances can be modeling using single- user data.
Bridging the Gap for Tokenizer-Free Language Models
- Dokook Choe, Rami Al-Rfou, Mandy Guo, Heeyoung Lee, Noah Constant
- Computer ScienceArXiv
- 27 August 2019
This paper trains a vanilla transformer network with 40 self-attention layers on the One Billion Word (lm1b) benchmark and achieves a new state of the art for tokenizer-free LMs, pushing these models to be on par with their word-based counterparts.
A Deterministic Coreference System with Rich Syntactic Features and Semantic Knowledge
This project has implemented and investigated numerous enhancements and feature extensions to the baseline system [Haghighi and Klein] part of JavaNLP 1, in addition to some important bug fixes.
Classifying Relationships Between Nouns
A system for classifying relationships between noun pairs by finding the dependency paths between the nouns in the Reuters corpus and training a multiclass classifier based on the counts of the paths achieves up to a 60% classification accuracy over four classes of word relationships using an SVM.