mlsanityData to 14 Oct 2026

Instructional Prompt Optimization for Few-Shot LLM-Based Recommendations on Cold-Start Users

arXiv 2509.09066 ↗First posted 11 Sep 2025cs.AIcited by 1

Not on the map. Why?

Abstract (arXiv)

The cold-start user issue further compromises the effectiveness of recommender systems in limiting access to the historical behavioral information. It is an effective pipeline to optimize instructional prompts on a few-shot large language model (LLM) used in recommender tasks. We introduce a context-conditioned prompt formulation method P(u, Ds) rightarrow Rwidehat, where u is a cold-start user profile, Ds is a curated support set, and Rwidehat is the predicted ranked list of items. Based on systematic experimentation with transformer-based autoregressive LLMs (BioGPT, LLaMA-2, GPT-4), we provide empirical evidence that optimal exemplar injection and instruction structuring can significantly improve the precision@k and NDCG scores of such models in low-data settings. The pipeline uses token-level alignments and embedding space regularization with a greater semantic fidelity. Our findings not only show that timely composition is not merely syntactic but also functional as it is in direct control of attention scales and decoder conduct through inference. This paper shows that prompt-based adaptation may be considered one of the ways to address cold-start recommendation issues in LLM-based pipelines.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 18 of its 30 references that are in the map. First posted on arXiv 11 Sep 2025.

Builds on

18 of 30 references.

References · 18

  1. 2024Enhancing Code LLMs with Reinforcement Learning in Code Generation: A Surveycited by 83
  2. 2024A Multimodal Fusion Network For Student Emotion Recognition Based on Transformer and Tensor Productcited by 22
  3. 2024Research on Splicing Image Detection Algorithms Based on Natural Image Statistical Characteristicscited by 21
  4. 2021Recurrent Neural Network from Adder's Perspective: Carry-lookahead RNNcited by 16
  5. 2024AR Overlay: Training Image Pose Estimation on Curved Surface in a Synthetic Waycited by 15
  6. 2024Research on reinforcement learning based warehouse robot navigation algorithm in complex warehouse layoutcited by 14
  7. 2024Enhanced Recommendation Combining Collaborative Filtering and Large Language Modelscited by 10
  8. 2024Rough Set improved Therapy-Based Metaverse Assisting Systemcited by 9
  9. 2025Research on the Design of a Short Video Recommendation System Based on Multimodal Information and Differential Privacycited by 9
  10. 2025Machine Learning-Based Prediction of Metal-Organic Framework Materials: A Comparative Analysis of Multiple Modelscited by 8
  11. 2025Machine Learning based Enterprise Financial Audit Framework and High Risk Identificationcited by 7
  12. 2025An automatic patent literature retrieval system based on LLM-RAGcited by 7
6 more
  1. 2025Research on Model Parallelism and Data Parallelism Optimization Methods in Large Language Model-Based Recommendation Systemscited by 6
  2. 2025A Deep Learning Algorithm Based on CNN-LSTM Framework for Predicting Cancer Drug Sales Volumecited by 5
  3. 2025Meta-Learning for Cold-Start Personalization in Prompt-Tuned LLMscited by 5
  4. 2025RLHF Fine-Tuning of LLMs for Alignment with Implicit User Feedback in Conversational Recommenderscited by 5
  5. 2025Deep Learning Model Acceleration and Optimization Strategies for Real-Time Recommendation Systemscited by 3
  6. 2025Research on Low-Latency Inference and Training Efficiency Optimization for Graph Neural Network and Large Language Model-Based Recommendation Systemscited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026RecPFN: Prior-Fitted Networks for In-Context-Based Recommendationscited by 0