# GenRecEdit: Adapting Model Editing for Generative Recommendation with Cold-Start Items

arXiv: https://arxiv.org/abs/2603.14259 · HTML: https://mlsanity.com/paper/2603.14259/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 24 references that are in the map.
- First posted on arXiv 15 Mar 2026.

## Builds on (most cited first, 20 of 20)

- [Transformer Feed-Forward Layers Are Key-Value Memories](https://mlsanity.com/paper/2012.14913/) (2020)
- [Inference-Time Intervention: Eliciting Truthful Answers from a Language Model](https://mlsanity.com/paper/2306.03341/) (2023)
- [Probing Classifiers: Promises, Shortcomings, and Advances](https://mlsanity.com/paper/2102.12452/) (2021)
- [Knowledge Neurons in Pretrained Transformers](https://mlsanity.com/paper/2104.08696/) (2021)
- [Recommender Systems with Generative Retrieval](https://mlsanity.com/paper/2305.05065/) (2023)
- [Learning Vector-Quantized Item Representation for Transferable Sequential Recommenders](https://mlsanity.com/paper/2210.12316/) (2022)
- [Learnable Item Tokenization for Generative Recommendation](https://mlsanity.com/paper/2405.07314/) (2024)
- [How to Index Item IDs for Recommendation Foundation Models](https://mlsanity.com/paper/2305.06569/) (2023)
- [EAGER: Two-Stream Generative Recommender with Behavior-Semantic Collaboration](https://mlsanity.com/paper/2406.14017/) (2024)
- [Generating Long Semantic IDs in Parallel for Recommendation](https://mlsanity.com/paper/2506.05781/) (2025)
- [CoST: Contrastive Quantization based Semantic Tokenization for Generative Recommendation](https://mlsanity.com/paper/2404.14774/) (2024)
- [Controllable Multi-Objective Re-ranking with Policy Hypernetworks](https://mlsanity.com/paper/2306.05118/) (2023)
- [Inductive Generative Recommendation via Retrieval-based Speculation](https://mlsanity.com/paper/2410.02939/) (2024)
- [Generative Retrieval with Semantic Tree-Structured Item Identifiers via Contrastive Learning](https://mlsanity.com/paper/2309.13375/) (2023)
- [A Survey of Controllable Learning: Methods and Applications in Information Retrieval](https://mlsanity.com/paper/2407.06083/) (2024)
- [HyperBandit: Contextual Bandit with Hypernewtork for Time-Varying User Preferences in Streaming Recommendation](https://mlsanity.com/paper/2308.08497/) (2023)
- [MoRE: A Mixture of Reflectors Framework for Large Language Model-Based Sequential Recommendation](https://mlsanity.com/paper/2409.06377/) (2024)
- [Paragon: Parameter Generation for Controllable Multi-Task Recommendation](https://mlsanity.com/paper/2410.10639/) (2024)
- [MAPS: Motivation-Aware Personalized Search via LLM-Driven Consultation Alignment](https://mlsanity.com/paper/2503.01711/) (2025)
- [Enhancing Bandit Algorithms with LLMs for Time-varying User Preferences in Streaming Recommendations](https://mlsanity.com/paper/2602.08067/) (2026)

## Built on by (most cited first, 1 of 1)

- [The Tokens Remember: When Tokenization Bypasses Knowledge Editing and Unlearning](https://mlsanity.com/paper/2609.29045/) (2026)

## Abstract (arXiv)

Generative recommendation (GR) has shown strong potential for sequential recommendation in an end-to-end generation paradigm. However, existing GR models suffer from severe cold-start collapse: their recommendation accuracy on cold-start items can drop to near zero. Current solutions typically rely on retraining with cold-start interactions, which is hindered by sparse feedback, high computational cost, and delayed updates, limiting practical utility in rapidly evolving recommendation catalogs. Inspired by model editing in NLP, which enables training-free knowledge injection into large language models, we explore how to bring this paradigm to generative recommendation. This, however, faces two key challenges: GR lacks the explicit subject-object binding common in natural language, making targeted edits difficult; and GR does not exhibit stable token co-occurrence patterns, making the injection of multi-token item representations unreliable. To address these challenges, we propose GenRecEdit, a model editing framework tailored for generative recommendation. GenRecEdit explicitly models the relationship between the full sequence context and next-token generation, adopts iterative token-level editing to inject multi-token item representations, and introduces a one-to-one trigger mechanism to reduce interference among multiple edits during inference. Extensive experiments on multiple datasets show that GenRecEdit substantially improves recommendation performance on cold-start items while preserving the model's original recommendation quality. Moreover, it achieves these gains using only about 9.5% of the training time required for retraining, enabling more efficient and frequent model updates.