AI Paper Atlas

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

Think Wise, Collaborate Effectively: A Rationale-Aware LLM-Based Recommender with Reinforcement Learning from Collaborative Signals

Chung Park, Taesan Kim, Hyeongjun Yun, Dongjoon Hong, Junui Hong, Kijung Park, MinCheol Cho, Min Sung Choi, Jihwan Seok, Jaegul Choo

AAAI 2026 · Large language models › Language, safety and interpretability · approach: LLM unlearning and benchmarks · 0 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 collaborative-guided reward (CF confidence) on top of item-matching and formatting rewards in GRPO (ablation)TWiCE-Rec · Amazon Musical Instruments and Health (HR@1, NDCG@5)HR@1, NDCG@5Improved
add Stage 2 instruction tuning before GRPO (ablation)TWiCE-Rec · Amazon Musical Instruments and HealthHR@1, NDCG@5Improved
add GRPO (Stage 3) after instruction tuning (ablation)TWiCE-Rec · Amazon Musical Instruments and HealthHR@1, NDCG@5Improved
apply RL only to the base model without Stage 2 (ablation)Gemma-3-1B-it · Amazon Musical Instruments and HealthHR@1, NDCG@5Mixed
increase CF confidence weight lambda up to 3 (ablation)TWiCE-Rec collaborative-guided reward · Amazon Fashion and ClothingHR@1, HR@5, NDCG@5Improved
increase lambda beyond 3 (ablation)TWiCE-Rec collaborative-guided reward · Amazon Fashion and ClothingHR@1, HR@5, NDCG@5Hurt
generate rationale at inference (ablation)TWiCE-Rec · eight Amazon domainsHR@1Improved
train with rationale supervision (ablation)TWiCE-Rec · eight Amazon domains (item-only inference)HR@1Improved
use TWiCE-Rec versus best baseline on HR@1TWiCE-Rec · Amazon FashionHR@1: +63.31%Improved
use TWiCE-Rec versus best baseline on HR@5TWiCE-Rec · Amazon Industrial and ScientificHR@5: 23.94 -> 23.16Hurt
use TWiCE-Rec in online serving versus ML modelTWiCE-Rec · online A/B testCTR: 10.67% -> 11.49%Improved
use larger LLM backbone (1B, 4B, 12B) (ablation)TWiCE-Rec · Amazon domainsrecommendation performanceImproved

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