mlsanityData to 7 Oct 2026

OneReason Technical Report

arXiv 2606.06260 ↗First posted 4 Jun 2026cs.IR cs.AI cs.CLcited by 8

On the map: LLM reasoning → LLM reasoning for recommendation

Abstract (arXiv)

Generative recommendation models in the OneRec family have been widely deployed in many real-world services, such as short-video, live-streaming, advertising, and e-commerce. However, these generative models can only benefit from the scaling advantage, while their reasoning ability is hard to activate, since we cannot construct meaningful Chain-of-Thought (CoT) sequences consisting of itemic tokens only. Inspired by the success of the reasoning-style ``think before answer'' paradigm in the LLM field, we conduct preliminary studies (i.e., OneRec-Think, OpenOneRec) to explore reasoning capability in generative recommendation. Nevertheless, we notice an unexpected phenomenon: the thinking mode does not show advantages over the non-thinking mode. Drawing insights from recent findings on CoT robustness in multi-modal language models, we argue that effective reasoning in recommendation rests on two factors: perception, the ability to ground itemic tokens in their underlying language semantics, and cognition, the ability to reorganize a user's behavior sequence into coherent latent interest points. We therefore propose OneReason, which includes: (1) strong itemic token perception in pre-training, (2) a three-level cognition-enhanced CoT format for recommendation tasks in SFT, and (3) a specialize-then-unify training recipe in RL to enhance the thinking ability.

Key facts. Cited by 8 papers in the map (data through 7 Oct 2026). Placed in LLM reasoning for recommendation, one of 17 directions in LLM reasoning. Builds on 59 of its 83 references that are in the map. First posted on arXiv 4 Jun 2026.

Builds on

59 of 83 references.

Closest ancestors · 6

Same direction, newest first

  1. 2026Reasoning over Semantic IDs Enhances Generative Recommendation24 Mar
  2. 2026Why Thinking Hurts: Diagnosing and Rectifying Linguistic Inertia in Large Language Models for Recommendation18 Feb
  3. 2025OxygenREC: An Instruction-Following Generative Framework for E-commerce Recommendation26 Dec
  4. 2025OneRec-Think: In-Text Reasoning for Generative Recommendation13 Oct
  5. 2025OnePiece: Bringing Context Engineering and Reasoning to Industrial Cascade Ranking System22 Sep
  6. 2025REG4Rec: Reasoning-Enhanced Generative Model for Large-Scale Recommendation Systems21 Aug

Other references · 53

  1. 2022Training language models to follow instructions with human feedbackcited by 15,712
  2. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
  3. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 8,851
  4. 2025Qwen3 Technical Reportcited by 8,195
  5. 2020Scaling Laws for Neural Language Modelscited by 6,411
  6. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,146
  7. 2022Large Language Models are Zero-Shot Reasonerscited by 4,323
  8. 2022Training Compute-Optimal Large Language Modelscited by 2,927
  9. 2025DAPO: An Open-Source LLM Reinforcement Learning System at Scalecited by 2,678
  10. 2024OpenAI o1 System Cardcited by 1,220
  11. 2025Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Modelscited by 1,175
  12. 2025Group Sequence Policy Optimizationcited by 788
38 more
  1. 2023Scaling Relationship on Learning Mathematical Reasoning with Large Language Modelscited by 417
  2. 2026Rethinking On-Policy Distillation of Large Language Models: Phenomenology, Mechanism, and Recipecited by 262
  3. 2023Chat-REC: Towards Interactive and Explainable LLMs-Augmented Recommender Systemcited by 241
  4. 2023Language Modeling Is Compressioncited by 225
  5. 2024Bridging Language and Items for Retrieval and Recommendation: Benchmarking LLMs as Semantic Encoderscited by 213
  6. 2024Actions Speak Louder than Words: Trillion-Parameter Sequential Transducers for Generative Recommendationscited by 184
  7. 2025OneRec: Unifying Retrieve and Rank with Generative Recommender and Iterative Preference Alignmentcited by 167
  8. 2026MiMo-V2-Flash Technical Reportcited by 163
  9. 2025MME-CoT: Benchmarking Chain-of-Thought in Large Multimodal Models for Reasoning Quality, Robustness, and Efficiencycited by 123
  10. 2023GPT4Rec: A Generative Framework for Personalized Recommendation and User Interests Interpretationcited by 101
  11. 2025Soft Adaptive Policy Optimizationcited by 89
  12. 2024Reuse, Don't Retrain: A Recipe for Continued Pretraining of Language Modelscited by 69
  13. 2021Counterfactual Explainable Recommendationcited by 60
  14. 2026Nemotron-Cascade 2: Post-Training LLMs with Cascade RL and Multi-Domain On-Policy Distillationcited by 53
  15. 2024InfoNCE: Identifying the Gap Between Theory and Practicecited by 39
  16. 2026Step 3.5 Flash: Open Frontier-Level Intelligence with 11B Active Parameterscited by 33
  17. 2025OneRec-V2 Technical Reportcited by 28
  18. 2025From Perception to Cognition: A Survey of Vision-Language Interactive Reasoning in Multimodal Large Language Modelscited by 26
  19. 2025OneRec Technical Reportcited by 24
  20. 2025OpenOneRec Technical Reportcited by 16
  21. 2026Counteraction-Aware Multi-Teacher On-Policy Distillation for General Capability Recovery with Domain Preservationcited by 14
  22. 2025Knowledge-Instruct: Effective Continual Pre-training from Limited Data using Instructionscited by 11
  23. 2025Corvid: Improving Multimodal Large Language Models Towards Chain-of-Thought Reasoningcited by 11
  24. 2025RecGPT Technical Reportcited by 11
  25. 2026Baichuan-M3: Modeling Clinical Inquiry for Reliable Medical Decision-Makingcited by 10
  26. 2025PinRec: Unified Generative Retrieval for Pinterest Recommender Systemscited by 8
  27. 2025RecGPT-V2 Technical Reportcited by 8
  28. 2026Generative Recommendation for Large-Scale Advertisingcited by 7
  29. 2026Reading, Not Thinking: Understanding and Bridging the Modality Gap When Text Becomes Pixels in Multimodal LLMscited by 6
  30. 2026OneSearch-V2: The Latent Reasoning Enhanced Self-distillation Generative Search Frameworkcited by 6
  31. 2025Reinforced Preference Optimization for Recommendationcited by 4
  32. 2026OneLive: Dynamically Unified Generative Framework for Live-Streaming Recommendationcited by 4
  33. 2026OneRanker: Unified Generation and Ranking with One Model in Industrial Advertising Recommendationcited by 4
  34. 2026Compression is all you need: Modeling Mathematicscited by 4
  35. 2026Fine-grained Semantics Integration for Large Language Model-based Recommendationcited by 3
  36. 2026How Well Does Generative Recommendation Generalize?cited by 3
  37. 2024LLM-Rank: A Graph Theoretical Approach to Pruning Large Language Modelscited by 2
  38. 2025An End-to-End Multi-objective Ensemble Ranking Framework for Video Recommendationcited by 2

50 most cited of 53 All 53 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 8.

8 papers.

  1. 2026Recommendation Retrievers Need Verifiers: Universal Generative Reranking for Sequential Recommendationscited by 1
  2. 2026WhisperRec: Latent Reasoning for Efficient Foundation Recommendation Modelscited by 0
  3. 2026Hierarchical Latent Reasoning for LLM-based Recommendationcited by 0
  4. 2026PersonaMem-v3: Toward Omni-Platform Personal Intelligence for Holistic User Understanding, Recommendation, and Agentic Taskscited by 0
  5. 2026The Disconnect Between Better Descriptive Reasoning Trace Quality and Recommendation Effectivenesscited by 0
  6. 2026T-RoPE: Time-Aware Rotary Position Embedding for Sequential Recommendationcited by 0
  7. 2026KUAISHOU Explorer LLM-Rec Challenge 2026: Reasoning Generative Recommendationcited by 0
  8. 2026Coverage-Aware Reasoning with Medical Tokens for Diagnosis Predictioncited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026LaRec: Unleashing LLM-based Latent Reasoning for Generative Recommendation
  2. 2026Implicit Reasoning for Large Language Model-based Generative Recommendation
  3. 2026S2GR: Stepwise Semantic-Guided Reasoning in Latent Space for Generative Recommendation
  4. 2026LASAR: Latent Adaptive Semantic Aligned Reasoning for Generative Recommendation
  5. 2025ThinkRec: Thinking-based recommendation via LLM
  6. 2026Learning Better Reasoning for Generative Recommendation with Semantic IDs
  7. 2025Generative Reasoning Recommendation via LLMs
  8. 2026TwiSTAR:Think Fast, Think Slow, Then Act,Generative Recommendation with Adaptive Reasoning