AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Hard vs. Noise: Resolving Hard-Noisy Sample Confusion in Recommender Systems via Large Language Models

Tianrui Song, Wen-Shuo Chao, Hao Liu

AAAI 2026 · Label-efficient and robust learning › Recognition and long-tail · approach: Recommendation representation learning · 4 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 LLM-embedding guided hard negative sampling (ablation)LLMHNI (LightGCN) · Amazon-bookRecall@10: 0.1199 -> 0.1423Improved
add objective alignment of LLM-encoded embeddings (ablation)LLMHNI (LightGCN) · Amazon-bookRecall@10: 0.1248 -> 0.1423Improved
add hallucination-robust graph contrastive loss (ablation)LLMHNI (LightGCN) · Amazon-bookRecall@10: 0.1125 -> 0.1423Improved
add cross-graph denoising contrastive loss (ablation)LLMHNI (LightGCN) · Amazon-bookRecall@10: 0.1294 -> 0.1423Improved
add hallucination-robust graph contrastive loss (ablation)LLMHNI (LightGCN) · Amazon-bookNDCG@10: 0.0855 -> 0.1168Improved
add hard negative sampling guided by semantic relevance (ablation)LLMHNI (LightGCN) · Amazon-bookNDCG@20: 0.1112 -> 0.1369Improved
use LLMHNI instead of LLaRD (LightGCN)LightGCN · Amazon-bookRecall@10: 0.1408 -> 0.1423Improved
use LLMHNI instead of vanilla trainingLightGCN · Amazon-bookRecall@10: 0.0950 -> 0.1423Improved
use LLMHNI instead of vanilla trainingNGCF · Amazon-bookRecall@10: 0.0763 -> 0.1290Improved
use LLMHNI instead of LLaRDLightGCN · YelpRecall@10: 0.0975 -> 0.0981Improved
add 5-20% non-existent training interactionsLLMHNI · Amazon-books and YelpRecallImproved
vary tau_hal and tau_de over 0.1-1.0LLMHNI · Amazon-booksRecall and NDCGMixed

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