mlsanityData to 7 Oct 2026

LLM reasoning for recommendation

160 papers0 this weekHigh confidence in this grouping: the papers form one clear topicHigh confidence in this grouping: the papers form one clear topic · About

Keywords: recommendation, llm-based, recommender, generative recommendation, generative

Key facts. LLM reasoning for recommendation: 160 papers on arXiv since 2022 (data through 7 Oct 2026). 19 of its 160 mapped papers (12%) cite “Leveraging LLM Reasoning Enhances Personalized Recommender Systems” (2024, arXiv 2408.00802). 19 of its 160 mapped papers (12%) cite “OneRec-Think: In-Text Reasoning for Generative Recommendation” (2025, arXiv 2510.11639). 2 new papers 17 Sep–7 Oct.

What’s new

17 Sep – 7 Oct 2026

2 new since 17 Sep

  1. Reasoning with Evidence, Not Merely Rationales: Verifiable Preference Proofs for LLM-Based Recommendation
  2. KUAISHOU Explorer LLM-Rec Challenge 2026: Reasoning Generative Recommendation

Foundations

Most cited by papers here

  1. Leveraging LLM Reasoning Enhances Personalized Recommender Systems2024cited by 19 here
  2. OneRec-Think: In-Text Reasoning for Generative Recommendation2025cited by 19 here
  3. Reinforced Latent Reasoning for LLM-based Recommendation2025cited by 18 here
  4. Reason4Rec: Deliberative User Preference Alignment of Large Language Models for Recommendation2025cited by 14 here
  5. Can Small Language Models be Good Reasoners for Sequential Recommendation?2024cited by 12 here

Lineage

The papers this direction builds on, oldest first.

Each paper on the line cites the one before.

  1. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 40 here
  2. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 12 here
  3. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 8 here
  4. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 12 here
  5. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 34 here
  6. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 33 here
  7. 2025Reinforced Latent Reasoning for LLM-based Recommendationcited by 18 here
  8. 2025OnePiece: Bringing Context Engineering and Reasoning to Industrial Cascade Ranking Systemcited by 9 here
16 more it builds on, not in this chain
  1. 2017Proximal Policy Optimization Algorithmscited by 15 here
  2. 2022Large Language Models are Zero-Shot Reasonerscited by 13 here
  3. 2023Tree of Thoughts: Deliberate Problem Solving with Large Language Modelscited by 9 here
  4. 2024Can Small Language Models be Good Reasoners for Sequential Recommendation?cited by 12 here
  5. 2024Leveraging LLM Reasoning Enhances Personalized Recommender Systemscited by 19 here
  6. 2024Review-driven Personalized Preference Reasoning with Large Language Models for Recommendationcited by 9 here
  7. 2024ReasoningRec: Bridging Personalized Recommendations and Human-Interpretable Explanations through LLM Reasoningcited by 9 here
  8. 2024DeepSeek-V3 Technical Reportcited by 9 here
  9. 2025Reason4Rec: Deliberative User Preference Alignment of Large Language Models for Recommendationcited by 14 here
  10. 2025DAPO: An Open-Source LLM Reinforcement Learning System at Scalecited by 12 here
  11. 2025Slow Thinking for Sequential Recommendationcited by 9 here
  12. 2025Qwen3 Technical Reportcited by 17 here
  13. 2025R2ec: Towards Large Recommender Models with Reasoningcited by 9 here
  14. 2025Reason-to-Recommend: Using Interaction-of-Thought Reasoning to Enhance LLM Recommendationcited by 12 here
  15. 2025R4ec: A Reasoning, Reflection, and Refinement Framework for Recommendation Systemscited by 9 here
  16. 2025OneRec-Think: In-Text Reasoning for Generative Recommendationcited by 19 here

The 24 most cited of 303.

Browse

30 of 160, ranked 6–35

  1. 6Reason-to-Recommend: Using Interaction-of-Thought Reasoning to Enhance LLM Recommendation202512
  2. 7Slow Thinking for Sequential Recommendation20259
  3. 8R2ec: Towards Large Recommender Models with Reasoning20259
  4. 9R4ec: A Reasoning, Reflection, and Refinement Framework for Recommendation Systems20259
  5. 10OnePiece: Bringing Context Engineering and Reasoning to Industrial Cascade Ranking System20259
  6. 11Review-driven Personalized Preference Reasoning with Large Language Models for Recommendation20249
  7. 12ReasoningRec: Bridging Personalized Recommendations and Human-Interpretable Explanations through LLM Reasoning20249
  8. 13Improving LLM-powered Recommendations with Personalized Information20257
  9. 14ThinkRec: Thinking-based recommendation via LLM20257
  10. 15Think before Recommendation: Autonomous Reasoning-enhanced Recommender20257
All 30
  1. 16Towards Next-Generation LLM-based Recommender Systems: A Survey and Beyond20247
  2. 17RecLLM-R1: A Two-Stage Training Paradigm with Reinforcement Learning and Chain-of-Thought v120255
  3. 18Rethinking Large Language Model Architectures for Sequential Recommendations20245
  4. 19Reasoning over Semantic IDs Enhances Generative Recommendation20264
  5. 20REG4Rec: Reasoning-Enhanced Generative Model for Large-Scale Recommendation Systems20254
  6. 21Generative Reasoning Recommendation via LLMs20254
  7. 22Large Language Models Enhanced Collaborative Filtering20244
  8. 23GOT4Rec: Graph of Thoughts for Sequential Recommendation20244
  9. 24Enhancing Recommender Systems with Large Language Model Reasoning Graphs20234
  10. 25S2GR: Stepwise Semantic-Guided Reasoning in Latent Space for Generative Recommendation20263
  11. 26GR2: Generative Reasoning Re-ranker20263
  12. 27Why Thinking Hurts: Diagnosing and Rectifying Linguistic Inertia in Large Language Models for Recommendation20263
  13. 28OneReason Technical Report20263
  14. 29Hierarchical Tree Search-based User Lifelong Behavior Modeling on Large Language Model20253
  15. 30OxygenREC: An Instruction-Following Generative Framework for E-commerce Recommendation20253
  16. 31Enhancing High-order Interaction Awareness in LLM-based Recommender Model20243
  17. 32Rethinking the Evaluation for Conversational Recommendation in the Era of Large Language Models20233
  18. 33DRDT: Dynamic Reflection with Divergent Thinking for LLM-based Sequential Recommendation20233
  19. 34Verifiable Reasoning for LLM-based Generative Recommendation20262
  20. 35A Unified Language Model for Large Scale Search, Recommendation, and Reasoning20262