HNHacker News
TopNewBestAskShowJobs

musculus

56 karma · joined November 22, 2025

submissionscomments

We train LLMs like dogs, not raise them: RLHF and sycophancy

old.reddit.com·1 pts·musculus·
6

Case study: Creative math – How AI fakes proofs

tomaszmachnik.pl·122 pts·musculus·
97

Reducing RLHF hallucinations and sycophancy in Gemini 3 (Interactive Demo)

tomaszmachnik.pl·1 pts·musculus·
0

RLHF Sycophancy: Gemini 3.0 discards calculated data to mimic user edits

tomaszmachnik.pl·3 pts·musculus·
0

Gemini 3.0 adopts user-injected hallucinations via history editing

tomaszmachnik.pl·1 pts·musculus·
0