cd/entity/Zhao et al.· home› entities› Zhao et al.
grep -l @zhao et al. /news/*.json | wc -l → 1

@Zhao et al.

mentions 1 type Person feed RSS
22:17
2026-05-26
lesswrong.com
large-language-models

Training Language Models for Controlled Stochasticity

Researchers have found that large language models exhibit severe bias and mode collapse when asked to generate random outputs, with models like Qwen3 selecting "Wednesday" 80% of the time when asked f…

// co-occurs with top 2 entities
Qwen3 1 Gemma-3 1