OneReason Technical Report

• URL: http://arxiv.org/abs/2606.06260v1
• Date: Thu, 04 Jun 2026 15:04:34 GMT
• Title: OneReason Technical Report
• Abstract summary: We show that generative recommendation models can only benefit from the scaling advantage, while their reasoning ability is hard to activate.<n>We 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.
• Score: 135.82017930661263
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: 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.

Related papers

• WhisperRec: Latent Reasoning for Efficient Foundation Recommendation Models [51.88075626243057]
  We propose WhisperRec, an efficient latent reasoning framework for foundation recommendation models (FRMs)<n>WhisperRec compresses teacher-generated Chain-of-Thought (CoT) into learnable latent reasoning tokens.<n>We show that WhisperRec consistently outperforms explicit-CoT methods and conventional baselines.
  arXiv  Detail & Related papers  (2026-07-29T08:48:35Z)
• Unlocking Fine-Grained Translation Quality Estimation in LRMs through Synergistically Evolving Implicit and Explicit Reasoning [40.63257408300584]
  RIEQE (Reasoning both Implicitly and Explicitly for QE) is a two-stage training framework that enables the co-evolution of implicit and explicit reasoning capabilities.<n>Results demonstrate that implicit and explicit reasoning co-evolve under our framework.
  arXiv  Detail & Related papers  (2026-05-29T14:47:49Z)
• When Is Thinking Enough? Early Exit via Sufficiency Assessment for Efficient Reasoning [52.21239821135325]
  We introduce Dynamic Thought Sufficiency in Reasoning (DTSR), a novel framework for efficient reasoning.<n>Inspired by human metacognition, DTSR operates in two stages: Reflection Signal Monitoring, and Thought Sufficiency Check.<n>DTSR reduces reasoning length by 28.9%-34.9% with minimal performance loss, effectively mitigating overthinking.
  arXiv  Detail & Related papers  (2026-04-08T07:56:28Z)
• Latent Chain-of-Thought for Visual Reasoning [53.541579327424046]
  Chain-of-thought (CoT) reasoning is critical for improving the interpretability and reliability of Large Vision-Language Models (LVLMs)<n>We reformulate reasoning in LVLMs as posterior inference and propose a scalable training algorithm based on amortized variational inference.<n>We empirically demonstrate that the proposed method enhances the state-of-the-art LVLMs on seven reasoning benchmarks.
  arXiv  Detail & Related papers  (2025-10-27T23:10:06Z)
• OneRec-Think: In-Text Reasoning for Generative Recommendation [55.53292983432484]
  OneRec-Think is a unified framework that seamlessly integrates dialogue, reasoning, and personalized recommendation.<n>Our proposed "Think-Ahead" architecture enables effective industrial deployment on Kuaishou, achieving a 0.159% gain in APP Stay Time.
  arXiv  Detail & Related papers  (2025-10-13T17:20:13Z)
• When Thinking Fails: The Pitfalls of Reasoning for Instruction-Following in LLMs [19.354141845315276]
  Chain-of-thought reasoning can significantly degrade instruction-following accuracy.<n>This is the first work to systematically expose reasoning-induced failures in instruction-following.
  arXiv  Detail & Related papers  (2025-05-16T16:36:00Z)
• Unified Multimodal Chain-of-Thought Reward Model through Reinforcement Fine-Tuning [31.727984223052648]
  This paper proposes UnifiedReward-Think, the first unified multimodal CoT-based reward model.<n>We first use a small amount of image generation preference data to distill the reasoning process of GPT-4o.<n>We then prepare large-scale unified multimodal preference data to elicit the model's reasoning process across various vision tasks.
  arXiv  Detail & Related papers  (2025-05-06T08:46:41Z)
• Efficient Inference for Large Reasoning Models: A Survey [74.17203483365171]
  Large Reasoning Models (LRMs) significantly improve the reasoning ability of Large Language Models (LLMs) by learning to reason.<n>However, their deliberative reasoning process leads to inefficiencies in token usage, memory consumption, and inference time.<n>This survey provides a review of efficient inference methods designed specifically for LRMs, focusing on mitigating token inefficiency while preserving the reasoning quality.
  arXiv  Detail & Related papers  (2025-03-29T13:27:46Z)
• Reason-RFT: Reinforcement Fine-Tuning for Visual Reasoning of Vision Language Models [42.75418134743927]
  Reason-RFT is a two-stage reinforcement fine-tuning framework for visual reasoning.<n>First,Supervised Fine-Tuning (SFT) with curated CoT data activates the reasoning potential of Vision-Language Models (VLMs)<n>Second, reinforcement learning based on Group Relative Policy Optimization (GRPO) generates multiple reasoning-response pairs to enhance adaptability to domain shifts.
  arXiv  Detail & Related papers  (2025-03-26T17:38:06Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.