AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

LLM-Aligned Geographic Item Tokenization for Local-Life Recommendation

Hao Jiang, Guoquan Wang, Donglin Zhou, Sheng Yu, Yang Zeng, Wencong Zeng, Kun Gai, Guorui Zhou

AAAI 2026 · Large language models › Language, safety and interpretability · approach: LLM unlearning and benchmarks · 6 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add LGSID tokenization to TWINTWIN · Kuaishou local-lifeAUC: 0.5898 -> 0.6263Improved
add LGSID tokenization to DIN vs RQ-VAEDIN · Kuaishou local-lifeAUC: 0.6185 -> 0.6276Improved
add LGSID tokenization to DIEN vs Res-KMeansDIEN · Kuaishou local-lifeAUC: 0.6369 -> 0.6484Improved
use LGSID instead of RQ-VAE in TIGERTIGER · Kuaishou local-lifeHit@10: 0.3880 -> 0.5077Improved
use LGSID instead of RQ-VAE in OneRecOneRec · Kuaishou local-lifeHit@10: 0.4534 -> 0.5537Improved
use Lin et al. quantization in TIGERTIGER · Kuaishou local-lifeHit@10: 0.3880 -> 0.2067Hurt
use list-wise reward instead of point-wise reward (ablation)G-DPO · Geographic alignment evalTown coverage T@5: 0.1452 -> 0.5584Improved
add density-aware list-wise sampling (ablation)G-DPO · Geographic alignment evalTown coverage T@5: 0.5584 -> 0.6114Improved
add domain mixed preference pairs (ablation)G-DPO · Geographic alignment evalSemantic similarity Top@5: 0.7411 -> 0.8107Improved
add domain mixed preference pairs (ablation)G-DPO · Geographic alignment evalTown coverage T@5: 0.6114 -> 0.1816Hurt
add similarity regularization (ablation)G-DPO · Geographic alignment evalSemantic similarity Top@5: 0.8107 -> 0.8856Improved
apply G-DPO alignment of the LLM (ablation)LGSID · Geographic alignment evalProvince coverage P@5: 0.8716 -> 0.9905Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy