Stanford CS329X HW2: 4 personalization methods via Thinking Machines' Tinker API
EchoShao8899 · x · 2026-11-04
Stanford CS329X HW2 lets students posttrain a production-scale LLM via Thinking Machines' Tinker API, which handles forward/backward passes so training fits on a laptop. Four personalization approaches to style-matched email generation: prompt engineering, SFT with synthetic data, RLAIF (reward model optimizing a personalized rubric), and DITTO for efficient online comparison data. Repo open-sourced.
仓库: https://github.com/SALT-NLP/cs329xhw2
More from Companies & People
- Claude Code PM Cat Wu leaves Anthropic, third top product voice out in weeks — alliekmiller · 2026-11-04
- Google says it removed post-employment non-disparagement clauses from new-hire terms — HaydnBelfield · 2026-11-04
- Accenture Accused of Publishing AI Paper Littered with AI Hallucinations — nordicinst · 2026-11-04
- Sarvam hate debate: critics vs its from-scratch training, bulbul rollout, enterprise work — AashaySachdeva · 2026-11-04
- Baidu's Apollo Go wraps two-week public robotaxi test rides in Tung Chung, Hong Kong — Baidu_Inc · 2026-11-04
- Founder of 50-person, 12-country team says he talks to AI agents more than humans — Scobleizer · 2026-11-04