# Recommendation Is a Dish Better Served Warm

arXiv: https://arxiv.org/abs/2508.07856 · HTML: https://mlsanity.com/paper/2508.07856/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 40 references that are in the map.
- First posted on arXiv 11 Aug 2025.

## Builds on (most cited first, 22 of 22)

- [Self-Attentive Sequential Recommendation](https://mlsanity.com/paper/1808.09781/) (2018)
- [Image-based Recommendations on Styles and Substitutes](https://mlsanity.com/paper/1506.04757/) (2015)
- [Are We Really Making Much Progress? A Worrying Analysis of Recent Neural Recommendation Approaches](https://mlsanity.com/paper/1907.06902/) (2019)
- [Embarrassingly Shallow Autoencoders for Sparse Data](https://mlsanity.com/paper/1905.03375/) (2019)
- [Denoising Implicit Feedback for Recommendation](https://mlsanity.com/paper/2006.04153/) (2020)
- [A Critical Study on Data Leakage in Recommender System Offline Evaluation](https://mlsanity.com/paper/2010.11060/) (2020)
- [Turning Dross Into Gold Loss: is BERT4Rec really better than SASRec?](https://mlsanity.com/paper/2309.07602/) (2023)
- [A Case Study on Sampling Strategies for Evaluating Neural Sequential Item Recommendation Models](https://mlsanity.com/paper/2107.13045/) (2021)
- [How Useful are Reviews for Recommendation? A Critical Review and Potential Improvements](https://mlsanity.com/paper/2005.12210/) (2020)
- [Widespread Flaws in Offline Evaluation of Recommender Systems](https://mlsanity.com/paper/2307.14951/) (2023)
- [Vista: A Visually, Socially, and Temporally-aware Model for Artistic Recommendation](https://mlsanity.com/paper/1607.04373/) (2016)
- [Deep Item-based Collaborative Filtering for Top-N Recommendation](https://mlsanity.com/paper/1811.04392/) (2018)
- [Take a Fresh Look at Recommender Systems from an Evaluation Standpoint](https://mlsanity.com/paper/2210.04149/) (2022)
- [Graph Collaborative Signals Denoising and Augmentation for Recommendation](https://mlsanity.com/paper/2304.03344/) (2023)
- [Scalable Cross-Entropy Loss for Sequential Recommendations with Large Item Catalogs](https://mlsanity.com/paper/2409.18721/) (2024)
- [Time to Split: Exploring Data Splitting Strategies for Offline Evaluation of Sequential Recommenders](https://mlsanity.com/paper/2507.16289/) (2025)
- [RECE: Reduced Cross-Entropy Loss for Large-Catalogue Sequential Recommenders](https://mlsanity.com/paper/2408.02354/) (2024)
- [HybridSVD: When Collaborative Information is Not Enough](https://mlsanity.com/paper/1802.06398/) (2018)
- [Autoregressive Generation Strategies for Top-K Sequential Recommendations](https://mlsanity.com/paper/2409.17730/) (2024)
- [Dynamic Modeling of User Preferences for Stable Recommendations](https://mlsanity.com/paper/2104.05047/) (2021)
- [The Cold-start Problem: Minimal Users' Activity Estimation](https://mlsanity.com/paper/2106.00102/) (2021)
- [Maximum Impact with Fewer Features: Efficient Feature Selection for Cold-Start Recommenders through Collaborative Importance Weighting](https://mlsanity.com/paper/2508.06455/) (2025)

## Built on by (most cited first, 2 of 2)

- [Adaptive Quality-Diversity Trade-offs for Large-Scale Batch Recommendation](https://mlsanity.com/paper/2602.02024/) (2026)
- [SplitLight: An Exploratory Toolkit for Recommender Systems Datasets and Splits](https://mlsanity.com/paper/2602.19339/) (2026)

## Abstract (arXiv)

In modern recommender systems, experimental settings typically include filtering out cold users and items based on a minimum interaction threshold. However, these thresholds are often chosen arbitrarily and vary widely across studies, leading to inconsistencies that can significantly affect the comparability and reliability of evaluation results. In this paper, we systematically explore the cold-start boundary by examining the criteria used to determine whether a user or an item should be considered cold. Our experiments incrementally vary the number of interactions for different items during training, and gradually update the length of user interaction histories during inference. We investigate the thresholds across several widely used datasets, commonly represented in recent papers from top-tier conferences, and on multiple established recommender baselines. Our findings show that inconsistent selection of cold-start thresholds can either result in the unnecessary removal of valuable data or lead to the misclassification of cold instances as warm, introducing more noise into the system.