Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
38 tokens/sec
GPT-4o
59 tokens/sec
Gemini 2.5 Pro Pro
41 tokens/sec
o3 Pro
7 tokens/sec
GPT-4.1 Pro
50 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

CoRAL: Collaborative Retrieval-Augmented Large Language Models Improve Long-tail Recommendation (2403.06447v1)

Published 11 Mar 2024 in cs.IR and cs.AI

Abstract: The long-tail recommendation is a challenging task for traditional recommender systems, due to data sparsity and data imbalance issues. The recent development of LLMs has shown their abilities in complex reasoning, which can help to deduce users' preferences based on very few previous interactions. However, since most LLM-based systems rely on items' semantic meaning as the sole evidence for reasoning, the collaborative information of user-item interactions is neglected, which can cause the LLM's reasoning to be misaligned with task-specific collaborative information of the dataset. To further align LLMs' reasoning to task-specific user-item interaction knowledge, we introduce collaborative retrieval-augmented LLMs, CoRAL, which directly incorporate collaborative evidence into the prompts. Based on the retrieved user-item interactions, the LLM can analyze shared and distinct preferences among users, and summarize the patterns indicating which types of users would be attracted by certain items. The retrieved collaborative evidence prompts the LLM to align its reasoning with the user-item interaction patterns in the dataset. However, since the capacity of the input prompt is limited, finding the minimally-sufficient collaborative information for recommendation tasks can be challenging. We propose to find the optimal interaction set through a sequential decision-making process and develop a retrieval policy learned through a reinforcement learning (RL) framework, CoRAL. Our experimental results show that CoRAL can significantly improve LLMs' reasoning abilities on specific recommendation tasks. Our analysis also reveals that CoRAL can more efficiently explore collaborative information through reinforcement learning.

PDF HTML Abstract
Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize File Document Download Save Streamline Icon: https://streamlinehq.com PDF File Document Download Save Streamline Icon: https://streamlinehq.com Markdown Bookmark Chat Bubble Oval Streamline Icon: https://streamlinehq.com Chat (Pro)
Definition Search Book Streamline Icon: https://streamlinehq.com
References (76)
  1. User-centered evaluation of popularity bias in recommender systems. In Proceedings of the 29th ACM Conference on User Modeling, Adaptation and Personalization. 119–129.
  2. Gpt-4 technical report. arXiv preprint arXiv:2303.08774 (2023).
  3. Reenvisioning the comparison between neural collaborative filtering and matrix factorization. In Proceedings of the 15th ACM Conference on Recommender Systems. 521–529.
  4. Knowledge-Augmented Large Language Models for Personalized Contextual Query Suggestion. arXiv preprint arXiv:2311.06318 (2023).
  5. Tallrec: An effective and efficient tuning framework to align large language model with recommendation. arXiv preprint arXiv:2305.00447 (2023).
  6. Stephen Bonner and Flavian Vasile. 2018. Causal embeddings for recommendation. In Proceedings of the 12th ACM conference on recommender systems. 104–112.
  7. Openai gym. arXiv preprint arXiv:1606.01540 (2016).
  8. Jonathon Byrd and Zachary Lipton. 2019. What is the effect of importance weighting in deep learning?. In International conference on machine learning. PMLR, 872–881.
  9. Efficient heterogeneous collaborative filtering without negative sampling for recommendation. In Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 34. 19–26.
  10. Wide & deep learning for recommender systems. In Proceedings of the 1st workshop on deep learning for recommender systems. 7–10.
  11. Class-balanced loss based on effective number of samples. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 9268–9277.
  12. Full Index Deep Retrieval: End-to-End User and Item Structures for Cold-start and Long-tail Item Recommendation. In Proceedings of the 17th ACM Conference on Recommender Systems. 47–57.
  13. Deep multifaceted transformers for multi-objective ranking in large-scale e-commerce recommender systems. In Proceedings of the 29th ACM International Conference on Information & Knowledge Management. 2493–2500.
  14. DeepFM: a factorization-machine based neural network for CTR prediction. arXiv preprint arXiv:1703.04247 (2017).
  15. Correcting exposure bias for link recommendation. In International Conference on Machine Learning. PMLR, 3953–3963.
  16. Leveraging large language models for sequential recommendation. In Proceedings of the 17th ACM Conference on Recommender Systems. 1096–1102.
  17. Do LLMs Understand User Preferences? Evaluating LLMs On User Rating Prediction. arXiv preprint arXiv:2305.06474 (2023).
  18. Sami Khenissi and Olfa Nasraoui. 2020. Modeling and counteracting exposure bias in recommender systems. arXiv preprint arXiv:2001.04832 (2020).
  19. Collaborative filtering based on collaborative tagging for enhancing the quality of recommendation. Electronic Commerce Research and Applications 9, 1 (2010), 73–83.
  20. FinePrompt: Unveiling the Role of Finetuned Inductive Bias on Compositional Reasoning in GPT-4. In Findings of the Association for Computational Linguistics: EMNLP 2023. 3763–3775.
  21. Diederik P Kingma and Jimmy Ba. 2014. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980 (2014).
  22. Theory of mind for multi-agent collaboration via large language models. arXiv preprint arXiv:2310.10701 (2023).
  23. Prompt distillation for efficient llm-based recommendation. In Proceedings of the 32nd ACM International Conference on Information and Knowledge Management. 1348–1357.
  24. Leave no user behind: Towards improving the utility of recommender systems for non-mainstream users. In Proceedings of the 14th ACM International Conference on Web Search and Data Mining. 103–111.
  25. Continuous control with deep reinforcement learning. arXiv preprint arXiv:1509.02971 (2015).
  26. Siyi Liu and Yujia Zheng. 2020. Long-tail session-based recommendation. In Proceedings of the 14th ACM Conference on Recommender Systems. 509–514.
  27. Interact with the Explanations: Causal Debiased Explainable Recommendation System. In Proceedings of the 17th ACM International Conference on Web Search and Data Mining. 472–481.
  28. Co-occurrence Embedding Enhancement for Long-tail Problem in Multi-Interest Recommendation. In Proceedings of the 17th ACM Conference on Recommender Systems. 820–825.
  29. Recommending long-tail items using extended tripartite graphs. In 2018 IEEE International Conference on Big Knowledge (ICBK). IEEE, 123–130.
  30. Improving Long-Tail Item Recommendation with Graph Augmentation. In Proceedings of the 32nd ACM International Conference on Information and Knowledge Management. 1707–1716.
  31. Large Language Models are Not Stable Recommender Systems. arXiv preprint arXiv:2312.15746 (2023).
  32. Long-tail learning via logit adjustment. arXiv preprint arXiv:2007.07314 (2020).
  33. Playing atari with deep reinforcement learning. arXiv preprint arXiv:1312.5602 (2013).
  34. Human-level control through deep reinforcement learning. nature 518, 7540 (2015), 529–533.
  35. Improving generative visual dialog by answering diverse questions. arXiv preprint arXiv:1909.10470 (2019).
  36. Justifying recommendations using distantly-labeled reviews and fine-grained aspects. In Proceedings of the 2019 conference on empirical methods in natural language processing and the 9th international joint conference on natural language processing (EMNLP-IJCNLP). 188–197.
  37. Correcting for selection bias in learning-to-rank systems. In Proceedings of The Web Conference 2020. 1863–1873.
  38. Grigorios A Pavliotis. 2016. Stochastic processes and applications. Springer.
  39. Stable-baselines3: Reliable reinforcement learning implementations. The Journal of Machine Learning Research 22, 1 (2021), 12348–12355.
  40. Experiments on generalizability of user-oriented fairness in recommender systems. In Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval. 2755–2764.
  41. Lkpnr: Llm and kg for personalized news recommendation framework. arXiv preprint arXiv:2308.12028 (2023).
  42. Large language models are competitive near cold-start recommenders for language-and item-based preferences. In Proceedings of the 17th ACM conference on recommender systems. 890–896.
  43. Recommendations as treatments: Debiasing learning and evaluation. In international conference on machine learning. PMLR, 1670–1679.
  44. Proximal policy optimization algorithms. arXiv preprint arXiv:1707.06347 (2017).
  45. Rama Syamala Sreepada and Bidyut Kr Patra. 2020. Mitigating long tail effect in recommendations using few shot learning technique. Expert Systems with Applications 140 (2020), 112887.
  46. Can ChatGPT Replace Traditional KBQA Models? An In-Depth Analysis of the Question Answering Performance of the GPT LLM Family. In International Semantic Web Conference. Springer, 348–367.
  47. Shuai Tang and Xiaofeng Zhang. 2021. CADPP: An Effective Approach to Recommend Attentive and Diverse Long-tail Items. In IEEE/WIC/ACM International Conference on Web Intelligence and Intelligent Agent Technology. 218–225.
  48. Boosting Language Models Reasoning with Chain-of-Knowledge Prompting. arXiv preprint arXiv:2306.06427 (2023).
  49. InstructGraph: Boosting Large Language Models via Graph-centric Instruction Tuning and Preference Alignment. arXiv preprint arXiv:2402.08785 (2024).
  50. Deep & cross network for ad click predictions. In Proceedings of the ADKDD’17. 1–7.
  51. Diffusion Recommender Model. arXiv preprint arXiv:2304.04971 (2023).
  52. Learning to rank with selection bias in personal search. In Proceedings of the 39th International ACM SIGIR conference on Research and Development in Information Retrieval. 115–124.
  53. Recmind: Large language model powered agent for recommendation. arXiv preprint arXiv:2308.14296 (2023).
  54. DRDT: Dynamic Reflection with Divergent Thinking for LLM-based Sequential Recommendation. arXiv preprint arXiv:2312.11336 (2023).
  55. Chain-of-thought prompting elicits reasoning in large language models. Advances in Neural Information Processing Systems 35 (2022), 24824–24837.
  56. Model-agnostic counterfactual reasoning for eliminating popularity bias in recommender system. In Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining. 1791–1800.
  57. Llmrec: Large language models with graph augmentation for recommendation. arXiv preprint arXiv:2311.00423 (2023).
  58. Dynamics-aware adaptation for reinforcement learning based cross-domain interactive recommendation. In Proceedings of the 45th International ACM SIGIR Conference on Research and Development in Information Retrieval. 290–300.
  59. Deconfounded and explainable interactive vision-language retrieval of complex scenes. In Proceedings of the 29th ACM International Conference on Multimedia. 2103–2111.
  60. User-regulation deconfounded conversational recommender system with bandit feedback. In Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. 2694–2704.
  61. Attentional factorization machines: Learning the weight of feature interactions via attention networks. arXiv preprint arXiv:1708.04617 (2017).
  62. Knowledge Plugins: Enhancing Large Language Models for Domain-Specific Recommendations. arXiv preprint arXiv:2311.10779 (2023).
  63. Sampling-bias-corrected neural modeling for large corpus item recommendations. In Proceedings of the 13th ACM Conference on Recommender Systems. 269–277.
  64. Challenging the long tail recommendation. arXiv preprint arXiv:1205.6700 (2012).
  65. Thought propagation: An analogical approach to complex reasoning with large language models. arXiv preprint arXiv:2310.03965 (2023).
  66. On the Long Tail Products Recommendation using Tripartite Graph. International Journal of Advanced Computer Science and Applications 13, 1 (2022).
  67. Fan Zhang and Qijie Shen. 2023. A Model-Agnostic Popularity Debias Training Framework for Click-Through Rate Prediction in Recommender System. In Proceedings of the 46th International ACM SIGIR Conference on Research and Development in Information Retrieval. 1760–1764.
  68. Robust Recommender System: A Survey and Future Directions. arXiv preprint arXiv:2309.02057 (2023).
  69. Bridging the Information Gap Between Domain-Specific Model and General LLM for Personalized Recommendation. arXiv preprint arXiv:2311.03778 (2023).
  70. A model of two tales: Dual transfer learning framework for improved long-tail item recommendation. In Proceedings of the web conference 2021. 2220–2231.
  71. Collm: Integrating collaborative embeddings into large language models for recommendation. arXiv preprint arXiv:2310.19488 (2023).
  72. Empowering Long-tail Item Recommendation through Cross Decoupling Network (CDN). In Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. 5608–5617.
  73. Automatic chain of thought prompting in large language models. arXiv preprint arXiv:2210.03493 (2022).
  74. Adapting large language models by integrating collaborative semantics for recommendation. arXiv preprint arXiv:2311.09049 (2023).
  75. Disentangling user interest and conformity for recommendation with causal embedding. In Proceedings of the Web Conference 2021. 2980–2991.
  76. Spatial-Temporal Aligned Multi-Agent Learning for Visual Dialog Systems. In Proceedings of the 30th ACM International Conference on Multimedia. 482–490.
User Edit Pencil Streamline Icon: https://streamlinehq.com
Authors (7)
  1. Junda Wu (35 papers)
  2. Cheng-Chun Chang (1 paper)
  3. Tong Yu (119 papers)
  4. Zhankui He (27 papers)
  5. Jianing Wang (50 papers)
  6. Yupeng Hou (33 papers)
  7. Julian McAuley (238 papers)
Citations (3)
View on Semantic Scholar
X Twitter Logo Streamline Icon: https://streamlinehq.com

Tweets