AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

The Global Anchor Method for Quantifying Linguistic Shifts and Domain Adaptation

Zi Yin, Vin Sachidananda, Balaji Prabhakar

NeurIPS 2018 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Conditional distribution alignment · 20 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
Use the global anchor distance instead of alignment to quantify corpus-level shiftsWord embeddings trained on separate corpora · Google Books, all pairs of years 1900–2000corpus-level shift detection: 0.826No clear effect
Compare embeddings of different dimensionalities using the global anchor methodGlobal anchor method · Embeddings with different dimensionalitiesapplicabilityImproved
Use matrix multiplications and parallel computation instead of constrained optimizationGlobal anchor method versus alignment · Corpus-level shift measurementimplementation simplicityImproved
Use distributional global anchors instead of frequency-based comparisonsLanguage evolution analysis · Google Books English fiction over timeresolution of linguistic shiftsImproved
Apply graph Laplacian embedding to global anchor distancesGlobal anchor method · Google Books English fiction over timerecovery of language-evolution trajectoryImproved
Apply graph Laplacian embedding to global anchor distancesGlobal anchor method · arXiv academic categoriesdomain clusteringImproved
Vary embedding dimensionalityAlignment/global anchor distance ratio · Embedding dimensionalities 100, 200, 400, and 500alignment/anchor distance ratio: 0.797, 0.811, 0.814, 0.809No clear effect

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