Trace

Xueyuan Zhao

individual

Technical AI safety researchAI safety meta

Received
$3,500
from 1 funders
Grants
1
Funders
1
Years active
2026

Grants over time

$0$1K$2K$3K$4K2026

Biggest funders

BlueDot Impact$3,500100%1

Grants received

DateFunderViaAmountCauseSourcePurpose
Jun 2026BlueDot Impact$3,500Technical AI safety research, Epistemic infrastructurebluedotTesting whether subliminal learning in model distillation is driven by shared unembedding/readout alignment, and whether this risk appears in LLMs.