PulseAugur
EN
LIVE 11:46:44

OneReason framework boosts generative recommendation model reasoning

Researchers have introduced OneReason, a new framework designed to enhance the reasoning capabilities of generative recommendation models. Existing models like the OneRec family struggle to activate reasoning due to the difficulty of creating meaningful item-only "Chain-of-Thought" sequences. OneReason addresses this by focusing on two key factors: perception, which grounds item tokens in their semantic meaning, and cognition, which reorganizes user behavior into interest points. The framework incorporates improved pre-training, a novel cognition-enhanced CoT format, and a specialized training recipe to boost reasoning. AI

IMPACT Enhances generative recommendation models by improving their reasoning capabilities, potentially leading to more relevant and personalized suggestions across various platforms.

RANK_REASON The cluster contains a technical report detailing a new research framework for generative recommendation models.

Read on arXiv cs.IR (Information Retrieval) →

AI-generated summary · Google Gemini · from 2 sources. How we write summaries →

COVERAGE [2]

  1. arXiv cs.CL TIER_1 English(EN) · OneRec Team, Biao Yang, Boyang Ding, Chenglong Chu, Dunju Zang, Fei Pan, Han Li, Hao Jiang, Honghui Bao, Huanjie Wang, Jian Liang, Jiangxia Cao, Jiao Ou, Jiaxin Deng, Jinghao Zhang, Kun Gai, Lu Ren, Peiru Du, Pengfei Zheng, Rongzhou Zhang, Ruiming Tang, … ·

    OneReason Technical Report

    arXiv:2606.06260v1 Announce Type: cross 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 …

  2. arXiv cs.IR (Information Retrieval) TIER_1 English(EN) · Ziyi Zhao ·

    OneReason Technical Report

    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 abili…