AI Paper Atlas

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

Amazon-M2: A Multilingual Multi-locale Shopping Session Dataset for Recommendation and Text Generation

Wei Jin, Haitao Mao, Zheng Li, Haoming Jiang, Chen Luo, Hongzhi Wen, Haoyu Han, Hanqing Lu, Zhengyang Wang, Ruirui Li, Zhen Li, Monica Xiao Cheng and 8 more

NeurIPS 2023 · Large language models › Language, safety and interpretability · approach: Multilingual LLM data and culture · 10 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
use popularity heuristic instead of deep session modelssession recommenders · Task 1 OverallMRR@100: 0.1757-0.2159 -> 0.2426Improved
use CORE instead of popularitysession recommenders · Task 1 OverallRecall@100: 0.4243 -> 0.5591Improved
pretrain on large locales and finetune on ES/FR/ITSRGNN · Task 2 OverallMRR@100: 0.2640 -> 0.2779Improved
pretrain and finetuneNARM · Task 2 OverallRecall@100: 0.5920 -> 0.6127Improved
pretrain and finetuneCORE · Task 2 OverallMRR@100: 0.2045 -> 0.1720Hurt
pretrain and finetuneSTAMP · Task 2 OverallRecall@100: 0.5217 -> 0.4803Hurt
initialize item embeddings with multilingual Sentence-BERT title embeddings (ablation)SRGNN · Task 2 OverallMRR@100: 0.2640 -> 0.2303Hurt
initialize item embeddings with Sentence-BERT title embeddings (ablation)GRU4Rec++ · Task 2 OverallRecall@100: 0.5856 -> 0.5454Hurt
increase session history length K from 1 to 3 (ablation)mT5-small · Task 3 phase-1 testBLEU: 0.2265 -> 0.2142No clear effect
increase model size from mT5-small to mT5-base (ablation)mT5, K=1 · Task 3 phase-1 testBLEU: 0.2265 -> 0.2251No clear effect
use Last Product Title heuristic instead of finetuned mT5title generation · Task 3 phase-1 testBLEU: 0.2265 -> 0.2677Improved

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