mlsanityData to 7 Oct 2026

Prototype-Aligned Federated Soft-Prompts for Continual Web Personalization

arXiv 2603.27678 ↗First posted 29 Mar 2026cs.LGcited by 5

Not on the map. Why?

Abstract (arXiv)

Continual web personalization is essential for engagement, yet real-world non-stationarity and privacy constraints make it hard to adapt quickly without forgetting long-term preferences. We target this gap by seeking a privacy-conscious, parameter-efficient interface that controls stability-plasticity at the user/session level while tying user memory to a shared semantic prior. We propose ProtoFed-SP, a prompt-based framework that injects dual-timescale soft prompts into a frozen backbone: a fast, sparse short-term prompt tracks session intent, while a slow long-term prompt is anchored to a small server-side prototype library that is continually refreshed via differentially private federated aggregation. Queries are routed to Top-M prototypes to compose a personalized prompt. Across eight benchmarks, ProtoFed-SP improves NDCG@10 by +2.9% and HR@10 by +2.0% over the strongest baselines, with notable gains on Amazon-Books (+5.0% NDCG vs. INFER), H&M (+2.5% vs. Dual-LoRA), and Taobao (+2.2% vs. FedRAP). It also lowers forgetting (AF) and Steps-to-95% and preserves accuracy under practical DP budgets. Our contribution is a unifying, privacy-aware prompting interface with prototype anchoring that delivers robust continual personalization and offers a transparent, controllable mechanism to balance stability and plasticity in deployment.

Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Builds on 20 of its 45 references that are in the map. First posted on arXiv 29 Mar 2026.

Builds on

20 of 45 references.

References · 20

  1. 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,592
  2. 2016Overcoming catastrophic forgetting in neural networkscited by 4,718
  3. 2019Parameter-Efficient Transfer Learning for NLPcited by 4,195
  4. 2021Prefix-Tuning: Optimizing Continuous Prompts for Generationcited by 3,878
  5. 2021The Power of Scale for Parameter-Efficient Prompt Tuningcited by 3,769
  6. 2018Federated Optimization in Heterogeneous Networkscited by 2,971
  7. 2016Learning without Forgettingcited by 2,465
  8. 2017Memory Aware Synapses: Learning what (not) to forgetcited by 1,094
  9. 2020Ditto: Fair and Robust Federated Learning Through Personalizationcited by 665
  10. 2024Large Language Models Empowered Personalized Web Agentscited by 64
  11. 2025Plan and Budget: Effective and Efficient Test-Time Scaling on Reasoning Large Language Modelscited by 30
  12. 2023Ensuring User-side Fairness in Dynamic Recommender Systemscited by 27
8 more
  1. 2024Swift Sampler: Efficient Learning of Sampler by 10 Parameterscited by 25
  2. 2024Post-Training Attribute Unlearning in Recommender Systemscited by 23
  3. 2024Cognitive Personalized Search Integrating Large Language Models with an Efficient Memory Mechanismcited by 22
  4. 2025LENSLLM: Unveiling Fine-Tuning Dynamics for LLM Selectioncited by 14
  5. 2022ReLoop: A Self-Correction Continual Learning Loop for Recommender Systemscited by 13
  6. 2023Structure Aware Incremental Learning with Personalized Imitation Weights for Recommender Systemscited by 9
  7. 2022Continual Transfer Learning for Cross-Domain Click-Through Rate Prediction at Taobaocited by 6
  8. 2024Real-Time Personalization for LLM-based Recommendation with Customized In-Context Learningcited by 5

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 5.

5 papers.

  1. 2026IMAGINE: Adaptive Schema-Imagery Enhanced Composition for Composed Video Retrievalcited by 1
  2. 2026MOSAIC: Orchestrating Collaborative Knowledge Tracing with Hierarchical Semantic Alignmentcited by 0
  3. 2026Foundation-Assisted Active Learning for Object Detection Annotationcited by 0
  4. 2026AsyncCouple-Flow: Asynchronous Cross-Modal Coupling and Flow Matching for Spatio-Temporal Forecastingcited by 0
  5. 2026CoRe: Coherence and Relational Alignment for Multivariate Time Series Forecastingcited by 0