Logo image
Back
LIMR: Intent-Aware Mashup API Recommendation via LLM-Augmented Multi-Scale Fusion
Journal article   Peer reviewed

LIMR: Intent-Aware Mashup API Recommendation via LLM-Augmented Multi-Scale Fusion

Yao Zhang, Yude Bai, Minhong Dong, Keqing Cen, Ji Zhang, Qiang Hu, Wei Ma, Yongqiang Lyu, Ruitao Feng, Xiaohong Li, …
IEEE transactions on services computing, Vol.First online(3), pp.1-16
2026
DOI: https://doi.org/10.1109/TSC.2026.3674665
Appears in  Recent Faculty of Science and Engineering Publications

Metrics

1 Record Views

Abstract

Mashups Semantics Grounding Cognition Accuracy Pattern matching Large language models Feature extraction Convolutional neural networks Computer architecture Mashup recommendation multi-scale model fusion large language model automated labeling
The increasing availability of Web APIs has amplified the complexity of mashup creation, where developers must identify compatible and functionally relevant APIs based on often ambiguous natural language descriptions. Traditional methods also fall short in capturing hierarchical semantic cues, modeling compatibility, and aligning with developer intent. Although large language models (LLMs) offer strong generalization capabilities, they remain unreliable in mashup recommendation due to hallucinated outputs, limited controllability, and token-length constraints when dealing with large-scale API repositories. To overcome these limitations, we introduce LIMR, an intent-aware mashup recommendation framework that combines LLM-augmented semantic reasoning with structured, multi-scale neural modeling. LIMR first prompts a LLM to extract high-level intent from user requirements, which serves as a global semantic signal. This intent is fused with low-level, multi-scale features extracted by a convolutional encoder, which are designed to capture fine-grained lexical/phrasal patterns at different granularities and provide precise semantic grounding for API matching. These heterogeneous representations are further contextually refined through a Transformer-based interaction module. To handle nonlinear semantic dependencies and compositional complexity, LIMR integrates a Kolmogorov-Arnold Network (KAN) with learnable activation functions, enhancing the model's capacity to capture intricate feature interactions. The entire framework is optimized via LLM, incorporating auxiliary objectives such as mashup category prediction and API quality estimation to guide generalization and reduce overfitting. Comprehensive experiments on the ProgrammableWeb and APIBench datasets show that LIMR significantly outperforms state-of-the-art baselines, which the ranking-oriented metrics, including NDCG and mAP, achieves improvements of 17.1%–34.2% over the strongest competitors. These results confirm the effectiveness of LIMR's hybrid design in delivering precise, robust, and intent-aware mashup API recommendations, especially in scenarios where LLMs alone fail to meet accuracy and scalability demands.

Details

Logo image