4 papers
PilotTTS: A Disciplined Modular Recipe for Competitive Speech Synthesis
Bowen Li, Shaotong Guo, Zhen Wang +11
Building state-of-the-art text-to-speech (TTS) systems typically demands millions of hours of proprietary data and complex multi-stage architectures, creating substantial barriers…
Rethinking Importance Sampling in LLM Policy Optimization: A Cumulative Token Perspective
Yuheng Zhang, Chenlu Ye, Shuowei Jin +4
Reinforcement learning, including reinforcement learning with verifiable rewards (RLVR), has emerged as a powerful approach for LLM post-training. Central to these approaches is th…
LLM Alignment as Retriever Optimization: An Information Retrieval Perspective
Bowen Jin, Jinsung Yoon, Zhen Qin +5
Large Language Models (LLMs) have revolutionized artificial intelligence with capabilities in reasoning, coding, and communication, driving innovation across industries. Their true…
RRM: Robust Reward Model Training Mitigates Reward Hacking
Tianqi Liu, Wei Xiong, Jie Ren +15
Reward models (RMs) play a pivotal role in aligning large language models (LLMs) with human preferences. However, traditional RM training, which relies on response pairs tied to sp…