# Sharpness-aware Model Merging with Salience Recovery for LLM-based Cross-Domain Sequential Recommendation

arXiv: https://arxiv.org/abs/2607.25366 · HTML: https://mlsanity.com/paper/2607.25366/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 31 of its 49 references that are in the map.
- First posted on arXiv 28 Jul 2026.

## Builds on (most cited first, 30 of 31)

- [Self-Attentive Sequential Recommendation](https://mlsanity.com/paper/1808.09781/) (2018)
- [Session-based Recommendations with Recurrent Neural Networks](https://mlsanity.com/paper/1511.06939/) (2015)
- [Recommendation as Language Processing (RLP): A Unified Pretrain, Personalized Prompt & Predict Paradigm (P5)](https://mlsanity.com/paper/2203.13366/) (2022)
- [M6-Rec: Generative Pretrained Language Models are Open-Ended Recommender Systems](https://mlsanity.com/paper/2205.08084/) (2022)
- [Animate-X: Universal Character Image Animation with Enhanced Motion Representation](https://mlsanity.com/paper/2410.10306/) (2024)
- [eCeLLM: Generalizing Large Language Models for E-commerce from Large-scale, High-quality Instruction Data](https://mlsanity.com/paper/2402.08831/) (2024)
- [Diffusion Augmentation for Sequential Recommendation](https://mlsanity.com/paper/2309.12858/) (2023)
- [DA-GCN: A Domain-aware Attentive Graph Convolution Network for Shared-account Cross-domain Sequential Recommendation](https://mlsanity.com/paper/2105.03300/) (2021)
- [M3oE: Multi-Domain Multi-Task Mixture-of Experts Recommendation Framework](https://mlsanity.com/paper/2404.18465/) (2024)
- [Contrastive Cross-Domain Sequential Recommendation](https://mlsanity.com/paper/2304.03891/) (2023)
- [Bridging Items and Language: A Transition Paradigm for Large Language Model-Based Recommendation](https://mlsanity.com/paper/2310.06491/) (2023)
- [ObjectRelator: Enabling Cross-View Object Relation Understanding Across Ego-Centric and Exo-Centric Perspectives](https://mlsanity.com/paper/2411.19083/) (2024)
- [Mixed Information Flow for Cross-domain Sequential Recommendations](https://mlsanity.com/paper/2012.00485/) (2020)
- [Parallel Split-Join Networks for Shared-account Cross-domain Sequential Recommendations](https://mlsanity.com/paper/1910.02448/) (2019)
- [Rethinking Cross-Domain Sequential Recommendation under Open-World Assumptions](https://mlsanity.com/paper/2311.04590/) (2023)
- [Exploring User Retrieval Integration towards Large Language Models for Cross-Domain Sequential Recommendation](https://mlsanity.com/paper/2406.03085/) (2024)
- [Dual Attentive Sequential Learning for Cross-Domain Click-Through Rate Prediction](https://mlsanity.com/paper/2106.02768/) (2021)
- [One Model for All: Large Language Models are Domain-Agnostic Recommendation Systems](https://mlsanity.com/paper/2310.14304/) (2023)
- [A Survey on Cross-Domain Sequential Recommendation](https://mlsanity.com/paper/2401.04971/) (2024)
- [Mimir: Improving Video Diffusion Models for Precise Text Understanding](https://mlsanity.com/paper/2412.03085/) (2024)
- [Mitigating Parameter Interference in Model Merging via Sharpness-Aware Fine-Tuning](https://mlsanity.com/paper/2504.14662/) (2025)
- [Bridge the Domains: Large Language Models Enhanced Cross-domain Sequential Recommendation](https://mlsanity.com/paper/2504.18383/) (2025)
- [VideoEraser: Concept Erasure in Text-to-Video Diffusion Models](https://mlsanity.com/paper/2508.15314/) (2025)
- [Pacer and Runner: Cooperative Learning Framework between Single- and Cross-Domain Sequential Recommendation](https://mlsanity.com/paper/2407.11245/) (2024)
- [Towards Open-world Cross-Domain Sequential Recommendation: A Model-Agnostic Contrastive Denoising Approach](https://mlsanity.com/paper/2311.04760/) (2023)
- [Uncovering Cross-Domain Recommendation Ability of Large Language Models](https://mlsanity.com/paper/2503.07761/) (2025)
- [Generalized Category Discovery under Domain Shift: A Frequency Domain Perspective](https://mlsanity.com/paper/2511.00573/) (2025)
- [WeaveRec: An LLM-Based Cross-Domain Sequential Recommendation Framework with Model Merging](https://mlsanity.com/paper/2510.26546/) (2025)
- [Cross Domain LifeLong Sequential Modeling for Online Click-Through Rate Prediction](https://mlsanity.com/paper/2312.06424/) (2023)
- [Cross-Domain Sequential Recommendation via Neural Process](https://mlsanity.com/paper/2410.13588/) (2024)

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

- [USA: Update-aware SAM for Cross-domain On-Policy Disitllation of Language Agents](https://mlsanity.com/paper/2609.34225/) (2026)

## Abstract (arXiv)

LLM-based Cross-Domain Sequential Recommendation (CDSR) leverages LLMs to enhance target performance via deep semantic reasoning, alleviating the dependency on overlapping users. Among LLM-based paradigms, model merging is particularly promising for multi-domain scenarios due to its superior scalability and flexibility in integrating diverse knowledge sources. However, our empirical investigations reveal two critical bottlenecks: (1) cross-domain knowledge conflict; and (2) performance saturation in multi-domain fusion. Our analysis attributes these phenomena to parameter-level misalignment and statistical homogenization during the merging process. To address these bottlenecks, we propose SharpRec, Sharpness-aware Model Merging with Salience Recovery for LLM-based CDSR, a framework designed to lift the performance upper bound of merged models. SharpRec incorporates two synergistic modules: Sharpness-aware Geometric Alignment to establish a stable geometric foundation for interference-free fusion; and Preference Salience Activation to effectively recover the distinctive features essential for bolstering target domain performance. Extensive experiments in both dual-domain and multi-domain scenarios demonstrate that SharpRec consistently outperforms state-of-the-art baselines.