Kuaishou's LIFT unifies retrieval and ranking, beats baselines by 4.9% on ML-20M
_reachsumit · x · 2026-10-09
Kuaishou researchers propose LIFT, unifying retrieval and ranking in recommender systems:
- Decomposes each interaction into ordered Request, Item, Context, and Action tokens modeled as a causal sequence.
- Retrieval reads the Request state while ranking reads the Context state, sharing one history encoder while preserving stage-specific information.
- Instantiated with Role-Conditioned Attention and a lightweight Pre-LN Bias.
On ML-20M and Taobao, LIFT achieves the highest Joint Score, beating the strongest baselines by 4.9% and 3.6%, with ablations and scaling analyses supporting each design choice.
More from Research
- RSI-Exam lands in State of AI report: top model Opus 5.5 scores just ~0.53 — cihangxie · 2026-10-09
- KAIST's ME-World tackles multi-agent egocentric world modeling with joint denoising — kaist-ai · 2026-10-09
- Controlled study: LLMs struggle to recover latent sequential structure despite long context — illinois · 2026-10-09
- TerraVis quantifies world-grounded visual consistency failures in text-to-image models — the-aiml · 2026-10-09
- OneSearch-VL: unified multimodal deep research agent beats Qwen3-VL by 20 points — Hongyu Li · 2026-10-09
- SpaceCast-Bench: best VLM hits 58.0% on predictive spatial reasoning vs 87.2% human — zju · 2026-10-09