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

arXiv: https://arxiv.org/abs/2511.07295 · HTML: https://mlsanity.com/paper/2511.07295/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 18 of its 23 references that are in the map.
- First posted on arXiv 10 Nov 2025.

## Builds on (most cited first, 18 of 18)

- [LightGCN: Simplifying and Powering Graph Convolution Network for Recommendation](https://mlsanity.com/paper/2002.02126/) (2020)
- [Neural Graph Collaborative Filtering](https://mlsanity.com/paper/1905.08108/) (2019)
- [A Survey on Large Language Models for Recommendation](https://mlsanity.com/paper/2305.19860/) (2023)
- [Self-supervised Graph Learning for Recommendation](https://mlsanity.com/paper/2010.10783/) (2020)
- [Adapting Large Language Models by Integrating Collaborative Semantics for Recommendation](https://mlsanity.com/paper/2311.09049/) (2023)
- [Are Graph Augmentations Necessary? Simple Graph Contrastive Learning for Recommendation](https://mlsanity.com/paper/2112.08679/) (2021)
- [LLMRec: Large Language Models with Graph Augmentation for Recommendation](https://mlsanity.com/paper/2311.00423/) (2023)
- [Representation Learning with Large Language Models for Recommendation](https://mlsanity.com/paper/2310.15950/) (2023)
- [XSimGCL: Towards Extremely Simple Graph Contrastive Learning for Recommendation](https://mlsanity.com/paper/2209.02544/) (2022)
- [Clicks can be Cheating: Counterfactual Recommendation for Mitigating Clickbait Issue](https://mlsanity.com/paper/2009.09945/) (2020)
- [Harnessing Large Language Models for Text-Rich Sequential Recommendation](https://mlsanity.com/paper/2403.13325/) (2024)
- [Simplify and Robustify Negative Sampling for Implicit Collaborative Filtering](https://mlsanity.com/paper/2009.03376/) (2020)
- [Self-Guided Learning to Denoise for Robust Recommendation](https://mlsanity.com/paper/2204.06832/) (2022)
- [AutoDenoise: Automatic Data Instance Denoising for Recommendations](https://mlsanity.com/paper/2303.06611/) (2023)
- [Efficient Bi-Level Optimization for Recommendation Denoising](https://mlsanity.com/paper/2210.10321/) (2022)
- [Double Correction Framework for Denoising Recommendation](https://mlsanity.com/paper/2405.11272/) (2024)
- [On the Theories Behind Hard Negative Sampling for Recommendation](https://mlsanity.com/paper/2302.03472/) (2023)
- [Spatial Object Recommendation with Hints: When Spatial Granularity Matters](https://mlsanity.com/paper/2101.02969/) (2021)

## Abstract (arXiv)

Implicit feedback, employed in training recommender systems, unavoidably confronts noise due to factors such as misclicks and position bias. Previous studies have attempted to identify noisy samples through their diverged data patterns, such as higher loss values, and mitigate their influence through sample dropping or reweighting. However, we observed that noisy samples and hard samples display similar patterns, leading to hard-noisy confusion issue. Such confusion is problematic as hard samples are vital for modeling user preferences. To solve this problem, we propose LLMHNI framework, leveraging two auxiliary user-item relevance signals generated by Large Language Models (LLMs) to differentiate hard and noisy samples. LLMHNI obtains user-item semantic relevance from LLM-encoded embeddings, which is used in negative sampling to select hard negatives while filtering out noisy false negatives. An objective alignment strategy is proposed to project LLM-encoded embeddings, originally for general language tasks, into a representation space optimized for user-item relevance modeling. LLMHNI also exploits LLM-inferred logical relevance within user-item interactions to identify hard and noisy samples. These LLM-inferred interactions are integrated into the interaction graph and guide denoising with cross-graph contrastive alignment. To eliminate the impact of unreliable interactions induced by LLM hallucination, we propose a graph contrastive learning strategy that aligns representations from randomly edge-dropped views to suppress unreliable edges. Empirical results demonstrate that LLMHNI significantly improves denoising and recommendation performance.