# OneSearch-V2: The Latent Reasoning Enhanced Self-distillation Generative Search Framework

arXiv: https://arxiv.org/abs/2603.24422 · HTML: https://mlsanity.com/paper/2603.24422/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Placed in On-policy and self-distillation for LLMs, one of 12 directions in Knowledge distillation.
- Builds on 24 of its 43 references that are in the map.
- First posted on arXiv 25 Mar 2026.

## Placed in

- Knowledge distillation / [On-policy and self-distillation for LLMs](https://mlsanity.com/distill/on-policy-and-self-distillation-for-llms/)

## Builds on (most cited first, 24 of 24)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension](https://mlsanity.com/paper/1910.13461/) (2019)
- [Training data-efficient image transformers & distillation through attention](https://mlsanity.com/paper/2012.12877/) (2020)
- [Qwen3-VL Technical Report](https://mlsanity.com/paper/2511.21631/) (2025)
- [Training Large Language Models to Reason in a Continuous Latent Space](https://mlsanity.com/paper/2412.06769/) (2024)
- [Deep Mutual Learning](https://mlsanity.com/paper/1706.00384/) (2017)
- [C-Pack: Packed Resources For General Chinese Embeddings](https://mlsanity.com/paper/2309.07597/) (2023)
- [Self-Distilled Reasoner: On-Policy Self-Distillation for Large Language Models](https://mlsanity.com/paper/2601.18734/) (2026)
- [Adversarial Training Methods for Semi-Supervised Text Classification](https://mlsanity.com/paper/1605.07725/) (2016)
- [CODI: Compressing Chain-of-Thought into Continuous Space via Self-Distillation](https://mlsanity.com/paper/2502.21074/) (2025)
- [Recommender Systems with Generative Retrieval](https://mlsanity.com/paper/2305.05065/) (2023)
- [OneRec: Unifying Retrieve and Rank with Generative Recommender and Iterative Preference Alignment](https://mlsanity.com/paper/2502.18965/) (2025)
- [IDGenRec: LLM-RecSys Alignment with Textual ID Learning](https://mlsanity.com/paper/2403.19021/) (2024)
- [Reinforced Latent Reasoning for LLM-based Recommendation](https://mlsanity.com/paper/2505.19092/) (2025)
- [OneRec-V2 Technical Report](https://mlsanity.com/paper/2508.20900/) (2025)
- [OneSearch: A Preliminary Exploration of the Unified End-to-End Generative Framework for E-commerce Search](https://mlsanity.com/paper/2509.03236/) (2025)
- [OneSug: The Unified End-to-End Generative Framework for E-commerce Query Suggestion](https://mlsanity.com/paper/2506.06913/) (2025)
- [Efficient Inference for Large Language Model-based Generative Recommendation](https://mlsanity.com/paper/2410.05165/) (2024)
- [Compositional Generalization from Learned Skills via CoT Training: A Theoretical and Structural Analysis for Reasoning](https://mlsanity.com/paper/2502.04667/) (2025)
- [Large Reasoning Embedding Models: Towards Next-Generation Dense Retrieval Paradigm](https://mlsanity.com/paper/2510.14321/) (2025)
- [EARN: Efficient Inference Acceleration for LLM-based Generative Recommendation by Register Tokens](https://mlsanity.com/paper/2507.00715/) (2025)
- [OneLoc: Geo-Aware Generative Recommender Systems for Local Life Service](https://mlsanity.com/paper/2508.14646/) (2025)
- [Listwise Direct Preference Optimization with Multi-Dimensional Preference Mixing](https://mlsanity.com/paper/2506.19780/) (2025)

## Built on by (most cited first, 6 of 6)

- [Preference-Based Self-Distillation: Beyond KL Matching via Reward Regularization](https://mlsanity.com/paper/2605.05040/) (2026)
- [OneReason Technical Report](https://mlsanity.com/paper/2606.06260/) (2026)
- [Bian Que: An Agentic Framework with Flexible Skill Arrangement for Online System Operations](https://mlsanity.com/paper/2604.26805/) (2026)
- [Plan Before Search: Search Agents Need Plan](https://mlsanity.com/paper/2605.28354/) (2026)
- [WhisperRec: Latent Reasoning for Efficient Foundation Recommendation Models](https://mlsanity.com/paper/2607.26621/) (2026)
- [UniPolicy: Unified Objective-Specific Policies for Generative Search Advertising](https://mlsanity.com/paper/2609.20630/) (2026)

## Abstract (arXiv)

Generative Retrieval (GR) has emerged as a promising paradigm for modern search systems. Compared to multi-stage cascaded architecture, it offers advantages such as end-to-end joint optimization and high computational efficiency. OneSearch, as a representative industrial-scale deployed generative search framework, has brought significant commercial and operational benefits. However, its inadequate understanding of complex queries, inefficient exploitation of latent user intents, and overfitting to narrow historical preferences have limited its further performance improvement. To address these challenges, we propose OneSearch-V2, a latent reasoning enhanced self-distillation generative search framework. It contains three key innovations: (1) a thought-augmented complex query understanding module, which enables deep query understanding and overcomes the shallow semantic matching limitations of direct inference; (2) a reasoning-internalized self-distillation training pipeline, which uncovers users' potential yet precise e-commerce intentions beyond log-fitting through implicit in-context learning; (3) a behavior preference alignment optimization system, which mitigates reward hacking arising from the single conversion metric, and addresses personal preference via direct user feedback. Extensive offline evaluations demonstrate OneSearch-V2's strong query recognition and user profiling capabilities. Online A/B tests further validate its business effectiveness, yielding +3.98% item CTR, +2.07% buyer volume, and +2.11% order volume. Manual evaluation further confirms gains in search experience quality, with +1.37% in page good rate and +1.65% in query-item relevance. More importantly, OneSearch-V2 effectively mitigates common search system issues such as information bubbles and long-tail sparsity, without incurring additional inference costs or serving latency.