解决大模型推荐中稳定、灵活与认知的三难困境。
TRACER: Balancing Stability-Plasticity-Cognitivity Trilemma for LLM Enhanced Continual Recommendation
- 设计三个模块分别应对稳定性、灵活性和认知性需求。
- 在五个数据集上提升效果，最高达14.38%。
- 适合需要持续学习推荐系统的开发者使用。
持续推荐旨在从流式数据中捕捉用户兴趣变化，但面临数据稀疏问题。大语言模型（LLM）通过引入语义知识缓解此问题，但直接整合引发新矛盾。我们识别出该矛盾为稳定性-灵活性-认知性（SPC）三难困境：通用的LLM语义先验（认知性）与保留个性化历史偏好（稳定性）及适应个体兴趣变化（灵活性）相冲突。为此，我们提出三难响应式自适应持续增强推荐框架（TRACER）。TRACER协同融合三个专用模块，分别聚焦稳定性、灵活性或认知性，防止任一模块主导。该整体设计使语义知识支持历史保留与兴趣演变适应，而不破坏持续学习过程。在五个真实世界数据集上，TRACER有效调和SPC三难困境，性能优于最先进基线最高达14.38%。代码已公开于 https://github.com/woo-joo/TRACER_CIKM26。
原文摘要 · Abstract (English)
Continual recommendation aims to capture evolving user interests from streaming data but struggles with sparsity. LLM enhancers mitigate this with semantic knowledge, but naive integration creates a new conflict. We identify this as the Stability-Plasticity-Cognitivity (SPC) Trilemma, where generalized LLM semantic priors (Cognitivity) conflict with retaining personalized historical preferences (Stability) and adapting to individual interest shifts (Plasticity). To address this, we propose Trilemma-Responsive Adaptive Continual Enhancement for Recommendation (TRACER). TRACER synergistically combines three specialized modules, each targeting stability, plasticity, or cognitivity, while preventing any single lemma from dominating. This holistic design enables semantic knowledge to support history retention and adaptation to evolving interests without disrupting continual learning. Across five real-world datasets, TRACER effectively harmonizes the SPC trilemma and outperforms state-of-the-art baselines by up to 14.38%. Our code is available at https://github.com/woo-joo/TRACER_CIKM26.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。