Only recent models seem to know "weight surgery" — are labs building Bayesian weight priors?
Tkaraletsos · x · 2026-11-03
Tkaraletsos observes that likely only recent models have been trained on enough sequences of weights to "know about" weight surgery — interventions like merging or transplanting model weights.
He quips, "Is everyone Bayesian again building weight priors?", hinting the field may be circling back to Bayesian-style weight priors. He frames this as the beginning of a trend, though the post offers no specifics on which models or evidence — an observational speculation.
More from Models
- Writing Weights Index: Open-Source Models Struggle at Weight-Space Programming — eliebakouch · 2026-11-03
- Gemini 4 Argon ships, putting Google back in the frontier model race after a year-long lull — danshipper · 2026-11-03
- Hands-on audit punctures Astra's '0.016% weight programming' claim: it just disabled Qwen's thinking mode — teortaxesTex · 2026-11-03
- Leak: Unreleased astra6.2 builds a 3DS-style Mario Kart in 20 min under 1% quota — lxfater · 2026-11-03
- "Sonnet 5.5 used to be smarter": users sense capability regression — nlarusstone · 2026-11-03
- Apple Intelligence lands in China in December, powered by Alibaba's Qwen instead of ChatGPT — winer666 · 2026-11-03