20 papers
Experience-Driven Dynamic Exits for LLMs with Reinforcement Learning
Yanyu Zhu, Hoilam Pao, Niu Hu +6
Large Language Models suffer from slow autoregressive inference. While self-speculative decoding accelerates this process, its efficiency is hampered by static configurations like…
TurboGR: An Accelerated Training System for Large-Scale Generative Recommendation
Huichao Chai, Zhixin Wu, Xuemiao Li +8
Generative recommendation (GR) has emerged as a promising paradigm that replaces fragmented, scenario-specific architectures with unified Transformer-based models, exhibiting scali…
Can Recommender Systems Teach Themselves? A Recursive Self-Improving Framework with Fidelity Control
Luankang Zhang, Hao Wang, Zhongzhou Liu +8
The scarcity of high-quality training data presents a fundamental bottleneck to scaling machine learning models. This challenge is particularly acute in recommendation systems, whe…
ReCast: Recasting Learning Signals for Reinforcement Learning in Generative Recommendation
Peiyan Zhang, Hanmo Liu, Chengxuan Tong +3
Generic group-based RL assumes that sampled rollout groups are already usable learning signals. We show that this assumption breaks down in sparse-hit generative recommendation, wh…
Generative Data Transformation: From Mixed to Unified Data
Jiaqing Zhang, Mingjia Yin, Hao Wang +6
Recommendation model performance is intrinsically tied to the quality, volume, and relevance of their training data. To address common challenges like data sparsity and cold start,…
SpecSteer: Synergizing Local Context and Global Reasoning for Efficient Personalized Generation
Hang Lv, Sheng Liang, Hao Wang +6
Realizing personalized intelligence faces a core dilemma: sending user history to centralized large language models raises privacy concerns, while on-device small language models l…