用轻量图模型实现学术冷启动推荐，速度提升50万倍，成本降99.9%。
TAG-HGT: A Scalable and Cost-Effective Framework for Inductive Cold-Start Academic Recommendation
- 先用大模型生成语义，再通过对比学习蒸馏到轻量图模型中
- 在OpenAlex上达SOTA召回率91.97%，比纯结构方法高20.7%
- 适合需要实时、低成本推荐的学术平台部署
学术冷启动推荐仍是工业级平台的难题，每日数千新学者缺乏历史交互数据。现有生成式图模型（如HiGPT、OFA）虽具良好语义能力，但推理延迟常超13分钟/千请求，计算成本过高，难以部署于百万级实时场景。为此，我们提出TAG-HGT，一种低成本神经符号框架。采用‘语义先行、结构精修’解耦策略，利用冻结的大型语言模型（DeepSeek-V3）作为离线语义工厂，通过跨视图对比学习（CVCL）将其知识蒸馏至轻量异质图变压器（HGT）。关键洞察：大模型提供全局召回，而结构信号则在稠密嵌入空间中区分有效合作者与语义相似但社交不可达的陌生人。在大规模OpenAlex数据集上，基于严格时间机器协议验证，TAG-HGT实现91.97%的SOTA系统召回率@10，较纯结构基线提升20.7%。从工业视角看，推理延迟降低五数量级（从780秒降至1.73毫秒），每千次查询成本由约1.50美元降至不足0.001美元，降幅达99.9%，真正实现高精度推荐的普惠化。
原文摘要 · Abstract (English)
Inductive cold-start recommendation remains the "Achilles' Heel" of industrial academic platforms, where thousands of new scholars join daily without historical interaction records. While recent Generative Graph Models (e.g., HiGPT, OFA) demonstrate promising semantic capabilities, their prohibitive inference latency (often exceeding 13 minutes per 1,000 requests) and massive computational costs render them practically undeployable for real-time, million-scale applications. To bridge this gap between generative quality and industrial scalability, we propose TAG-HGT, a cost-effective neuro-symbolic framework. Adopting a decoupled "Semantics-First, Structure-Refined" paradigm, TAG-HGT utilizes a frozen Large Language Model (DeepSeek-V3) as an offline semantic factory and distills its knowledge into a lightweight Heterogeneous Graph Transformer (HGT) via Cross-View Contrastive Learning (CVCL). We present a key insight: while LLM semantics provide necessary global recall, structural signals offer the critical local discrimination needed to distinguish valid collaborators from semantically similar but socially unreachable strangers in dense embedding spaces. Validated under a strict Time-Machine Protocol on the massive OpenAlex dataset, TAG-HGT achieves a SOTA System Recall@10 of 91.97%, outperforming structure-only baselines by 20.7%. Most significantly, from an industrial perspective, TAG-HGT reduces inference latency by five orders of magnitude ($4.5 \times 10^{5}\times$) compared to generative baselines (from 780s down to 1.73 ms), and slashes inference costs from $\sim$$1.50 to $<$$0.001 per 1k queries. This 99.9% cost reduction democratizes high-precision academic recommendation.
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。