GitHub / pixiiidust / semantic-ising
Do semantically identical words across languages converge in embedding space? Currently WIP and testing: This tool aims to visualize multilingual embedding alignment under Ising dynamics, revealing latent structure as the system approaches critical temperature. Inspired by the Platonic representation hypothesis.
JSON API: http://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pixiiidust%2Fsemantic-ising
PURL: pkg:github/pixiiidust/semantic-ising
Stars: 0
Forks: 0
Open issues: 0
License: None
Language: Python
Size: 548 KB
Dependencies parsed at: Pending
Created at: 3 months ago
Updated at: 3 months ago
Pushed at: 3 months ago
Last synced at: 3 months ago
Topics: ising-model, labse, multilingual-models, multilingual-translations, representation-learning, sbert, world-model, xlm-roberta