DoorDash's PALATE: GPT-style user embeddings lift ad revenue ~1% in production
_reachsumit · x · 2026-10-30
DoorDash's PALATE paper trains a GPT-style model on consumer store-engagement sequences and uses the last-token hidden state as a frozen per-consumer embedding shared across ad rankers. An exploration-weighted loss up-weights new stores for repurchase-heavy marketplaces. Offline it was their largest single-feature ranking-AUC gain (+0.570.59% relative); online it shipped with revenue +0.79% on Discovery and +0.94% on Search, with conversion up 1.17% and 1.50%.
More from Research
- OpenFlyV2: a 1B onboard VLA model for real-time aerial drone navigation — IPEC-COMMUNITY · 2026-10-30
- Prime gap bound shrinks from 70M to 212 since Zhang's breakthrough; twin primes needs 2 — IgorCarron · 2026-10-30
- Researcher: concept neurons alone may not be enough to induce an AI persona — Sauers_ · 2026-10-30
- Sibling model neuron sharing: probing shared representations across related models — Sauers_ · 2026-10-30
- Porting PCRE2 to Verified Rust: Fuzz Aggressively Using the C Version as an Oracle — jd_pressman · 2026-10-30
- UCLouvain Releases Full Video Lectures: Stochastic Optimization and Energy Economics — caglar_ee · 2026-10-30