# RecExplainer: Aligning Large Language Models for Explaining Recommendation Models

arXiv: https://arxiv.org/abs/2311.10947 · HTML: https://mlsanity.com/paper/2311.10947/

## Key facts

- Cited by 14 papers in the map (data through 14 Oct 2026).
- Placed in Preference alignment for recommenders and RAG, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 26 of its 52 references that are in the map.
- First posted on arXiv 18 Nov 2023.

## Placed in

- LLM alignment / RLHF / preference optimisation / [Preference alignment for recommenders and RAG](https://mlsanity.com/alignment/preference-alignment-for-recommenders-and-rag/)

## Builds on (most cited first, 26 of 26)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [A Unified Approach to Interpreting Model Predictions](https://mlsanity.com/paper/1705.07874/) (2017)
- ["Why Should I Trust You?": Explaining the Predictions of Any Classifier](https://mlsanity.com/paper/1602.04938/) (2016)
- [Robust Speech Recognition via Large-Scale Weak Supervision](https://mlsanity.com/paper/2212.04356/) (2022)
- [OPT: Open Pre-trained Transformer Language Models](https://mlsanity.com/paper/2205.01068/) (2022)
- [Visualizing and Understanding Convolutional Networks](https://mlsanity.com/paper/1311.2901/) (2013)
- [Understanding Black-box Predictions via Influence Functions](https://mlsanity.com/paper/1703.04730/) (2017)
- [Learning Important Features Through Propagating Activation Differences](https://mlsanity.com/paper/1704.02685/) (2017)
- [ZeRO: Memory Optimizations Toward Training Trillion Parameter Models](https://mlsanity.com/paper/1910.02054/) (2019)
- [VisualBERT: A Simple and Performant Baseline for Vision and Language](https://mlsanity.com/paper/1908.03557/) (2019)
- [Self-Attentive Sequential Recommendation](https://mlsanity.com/paper/1808.09781/) (2018)
- [WizardLM: Empowering large pre-trained language models to follow complex instructions](https://mlsanity.com/paper/2304.12244/) (2023)
- [GNNExplainer: Generating Explanations for Graph Neural Networks](https://mlsanity.com/paper/1903.03894/) (2019)
- [Can Large Language Models Be an Alternative to Human Evaluations?](https://mlsanity.com/paper/2305.01937/) (2023)
- [Unified Language Model Pre-training for Natural Language Understanding and Generation](https://mlsanity.com/paper/1905.03197/) (2019)
- [Textbooks Are All You Need II: phi-1.5 technical report](https://mlsanity.com/paper/2309.05463/) (2023)
- [Language Is Not All You Need: Aligning Perception with Language Models](https://mlsanity.com/paper/2302.14045/) (2023)
- [Understanding Neural Networks Through Deep Visualization](https://mlsanity.com/paper/1506.06579/) (2015)
- [Learning to Explain: An Information-Theoretic Perspective on Model Interpretation](https://mlsanity.com/paper/1802.07814/) (2018)
- [Is GPT-3 a Good Data Annotator?](https://mlsanity.com/paper/2212.10450/) (2022)
- [Recommendation as Instruction Following: A Large Language Model Empowered Recommendation Approach](https://mlsanity.com/paper/2305.07001/) (2023)
- [VLMixer: Unpaired Vision-Language Pre-training via Cross-Modal CutMix](https://mlsanity.com/paper/2206.08919/) (2022)

## Built on by (most cited first, 14 of 14)

- [Explaining Length Bias in LLM-Based Preference Evaluations](https://mlsanity.com/paper/2407.01085/) (2024)
- [Language Representations Can be What Recommenders Need: Findings and Potentials](https://mlsanity.com/paper/2407.05441/) (2024)
- [Improving LLM-powered Recommendations with Personalized Information](https://mlsanity.com/paper/2502.13845/) (2025)
- [Advancing and Benchmarking Personalized Tool Invocation for LLMs](https://mlsanity.com/paper/2505.04072/) (2025)
- [Explainable CTR Prediction via LLM Reasoning](https://mlsanity.com/paper/2412.02588/) (2024)
- [Tuning Language Models for Robust Prediction of Diverse User Behaviors](https://mlsanity.com/paper/2505.17682/) (2025)
- [Retrieval-Augmented Recommendation Explanation Generation with Hierarchical Aggregation](https://mlsanity.com/paper/2507.09188/) (2025)
- [HoloRec: Holistic Encoding and Interleaved Reasoning for Generative Recommendation](https://mlsanity.com/paper/2606.15331/) (2026)
- [Unveiling the Learning Mind of Language Models: A Cognitive Framework and Empirical Study](https://mlsanity.com/paper/2506.13464/) (2025)
- [Towards Trustworthy LLM-Based Recommendation via Rationale Integration](https://mlsanity.com/paper/2601.02364/) (2025)
- [From latent factors to language: a user study on LLM-generated explanations for an inherently interpretable matrix-based recommender system](https://mlsanity.com/paper/2509.18980/) (2025)
- [Interactive AI NPCs Powered by LLMs: Technical Report for the CPDC Challenge 2025](https://mlsanity.com/paper/2511.20200/) (2025)
- [LLMs Reading the Rhythms of Daily Life: Aligned Understanding for Behavior Prediction and Generation](https://mlsanity.com/paper/2604.23578/) (2026)
- [The Utility of LLMs in Recommender Systems Explanation Evaluation](https://mlsanity.com/paper/2609.01627/) (2026)

## Abstract (arXiv)

Recommender systems are widely used in online services, with embedding-based models being particularly popular due to their expressiveness in representing complex signals. However, these models often function as a black box, making them less transparent and reliable for both users and developers. Recently, large language models (LLMs) have demonstrated remarkable intelligence in understanding, reasoning, and instruction following. This paper presents the initial exploration of using LLMs as surrogate models to explaining black-box recommender models. The primary concept involves training LLMs to comprehend and emulate the behavior of target recommender models. By leveraging LLMs' own extensive world knowledge and multi-step reasoning abilities, these aligned LLMs can serve as advanced surrogates, capable of reasoning about observations. Moreover, employing natural language as an interface allows for the creation of customizable explanations that can be adapted to individual user preferences. To facilitate an effective alignment, we introduce three methods: behavior alignment, intention alignment, and hybrid alignment. Behavior alignment operates in the language space, representing user preferences and item information as text to mimic the target model's behavior; intention alignment works in the latent space of the recommendation model, using user and item representations to understand the model's behavior; hybrid alignment combines both language and latent spaces. Comprehensive experiments conducted on three public datasets show that our approach yields promising results in understanding and mimicking target models, producing high-quality, high-fidelity, and distinct explanations. Our code is available at https://github.com/microsoft/RecAI.