AI Paper Atlas

Research map / Learning theory and algorithms / Submodular and game algorithms

Greedy Flipping for Constrained Word Deletion

Jin-ge Yao, Xiaojun Wan

AAAI 2017 · Learning theory and algorithms › Submodular and game algorithms · approach: Streaming submodular algorithms · 2 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace ILP solver with randomized constrained greedy flippingLM+Disc objective · Written corpusinference time per sentence (s): 1.72 -> 0.28Improved
replace ILP solver with randomized constrained greedy flippingLM+Disc objective · Spoken corpusinference time per sentence (s): 1.49 -> 0.25Improved
replace ILP solver with randomized constrained greedy flippingLM+Disc objective · Written corpusGR-F1 (%): 66.5 -> 66.4No clear effect
replace ILP solver with randomized constrained greedy flippingLM+Disc objective · Spoken corpusGR-F1 (%): 66.2 -> 66.2No clear effect
replace ILP solver with randomized constrained greedy flippingLM objective · Written corpusinference time per sentence (s): 1.23 -> 0.25Improved
replace ILP solver with randomized constrained greedy flippingDisc objective · Spoken corpusGR-F1 (%): 56.2 -> 55.9No clear effect
add discriminative scores to language model scores in the objective (ablation)RCGF · Written corpusGR-F1 (%): 64.5 -> 66.4Improved
add discriminative scores to language model scores in the objective (ablation)ILP · Spoken corpusGR-F1 (%): 61.5 -> 66.2Improved
use only discriminative scores instead of language model scores (ablation)ILP · Written corpusGR-F1 (%): 64.5 -> 62.0Hurt
add explicit syntactic constraints (ILP/RCGF) versus unconstrained baselinesRCGF(LM+Disc) · Written corpusGR-F1 (%): 52.4 -> 66.4Improved
use RCGF instead of ILP in human evaluationLM+Disc · Written corpus manual ratinggrammaticality (1-5): 4.56 -> 4.56No clear effect
add explicit constraints versus McDonald06 in human evaluationRCGF(LM+Disc) · Spoken corpus manual ratinggrammaticality (1-5): 3.77 -> 4.38Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy