Inherit4Rec lets recommendation models scale up or convert to sparse experts by inheriting pretrained parameters
_reachsumit · x · 2026-09-22
A paper proposes Inherit4Rec, a parameter-inheritance framework for efficiently scaling industrial recommender models.
- Pain point: repeatedly training larger dense recommenders from scratch is costly, and their compute conflicts with strict serving budgets.
- Design: Dense-to-Dense (D2D) growth uses hybrid expansion with asymmetric training to preserve forward function and update continuity; Dense-to-Sparse (D2S) builds SMoE networks via co-activation-aware partitioning and a load-balancing loss.
- Results: on KuaiRand-1K and an industrial short-video recommendation dataset, both transformations consistently outperform inheritance baselines across all prediction objectives.
More from Research
- Tencent ARC Lab releases GameHorizon Suite to benchmark VLMs and agents on AAA gameplay — yshan2u · 2026-09-22
- Stanford turns papers into AI agents; two unrelated studies surface unreported ADHD genetic link — VraserX · 2026-09-22
- Xiaomi MiMo 2.6 Pro formalizes Li–Yorke chaos theorem in 6,000+ lines of verified Lean — bookwormengr · 2026-09-22
- Apollo, the first advanced LLM for Ancient Greek, aims to restore tattered papyri — nordicinst · 2026-09-22
- Two Minute Papers: Jev claims 200x AI speedup via System One models — but there's a catch — Two Minute Papers · 2026-09-22
- New LLM fills gaps in tattered ancient Greek papyrus records — Wired AI · 2026-09-22