GAVAGAI paper proposes balanced optimal transport for referential alignment in infant-scale vision-language models
A new arXiv paper, GAVAGAI, addresses how infant-scale vision-language models bind words to objects, using balanced optimal transport to achieve referential alignment from roughly 100 million words of child-like input.
Read the full story at arXiv cs.CL ↗
Timeline · 1 report
- 2026-11-03 05:00 · arXiv cs.CL
GAVAGAI: Referential Alignment for Infant-Scale Vision-Language Models via Balanced Optimal Transport