Alibaba's TriGen generative recommender lifts revenue across six markets in A/B tests
_reachsumit · x · 2026-11-13
Alibaba's TriGen targets Trigger-Induced Recommendation (TIR), where a click opens an immediate follow-up feed that must align with instant intent without breaking long-term preference continuity. It uses calibrated tail injection to re-encode local history, dynamic gating to arbitrate persistent tastes vs. immediate impulses, and position-specific transformations for level-aware trigger guidance. Offline evals beat competitive baselines, online A/B tests across six country markets showed significant revenue and GMV gains, and TriGen is now fully deployed in production.
More from Research
- François Fleuret asks: is the gradient Japanese? It never refuses to give you a direction — francoisfleuret · 2026-11-22
- Hat Matrix Explained: How Much Leverage Does Each Data Point Have in Least Squares? — RobertTLange · 2026-11-22
- Energy-Based Reasoners: gradient descent on latent states for adaptive reasoning — kalyan_kpl · 2026-11-22
- Lean community certifies Riemann proof; honeypot fake keys expose breaches at three AI providers — LeviTurk · 2026-11-22
- Riemann hypothesis 'proven by Turkish mathematician using AI' claim mocked as the 'Organic Turk' — CatAstro_Piyush · 2026-11-22
- Andriy Burkov releases The Hundred-Page Language Models Book, hands-on with PyTorch — burkov · 2026-11-22