27 papers
Hierarchical Acoustic-Semantic Modeling: Modality Separation and Semantic Coherence for Full-Duplex SLMs
Zhenyu Liu, Xuanyu Zhang, Yunxin Li +10
The paper identifies gradient conflicts between acoustic and semantic modeling as the cause of modality interference in full-duplex spoken language models and proposes Lychee-FD, a…
The Mirage of Optimizing Training Policies: Monotonic Inference Policies as the Real Objective for LLM Reinforcement Learning
Jing Liang, Hongyao Tang, Yi Ma +9
Reinforcement learning (RL) has gained growing attention in large language model (LLM) post-training, yet RL training remains fragile and can suffer from instability or collapse. O…
CodeTracer: Towards Traceable Agent States
Han Li, Yifan Yao, Letian Zhu +13
Code agents are advancing rapidly, but debugging them is becoming increasingly difficult. As frameworks orchestrate parallel tool calls and multi-stage workflows over complex tasks…
Long-form RewardBench: Evaluating Reward Models for Long-form Generation
Hui Huang, Yancheng He, Wei Liu +7
The widespread adoption of reinforcement learning-based alignment highlights the growing importance of reward models. Various benchmarks have been built to evaluate reward models i…
Let It Flow: Agentic Crafting on Rock and Roll, Building the ROME Model within an Open Agentic Learning Ecosystem
Weixun Wang, XiaoXiao Xu, Wanhe An +86
Agentic crafting requires LLMs to operate in real-world environments over multiple turns by taking actions, observing outcomes, and iteratively refining artifacts. Despite its impo…
CE-RM: A Pointwise Generative Reward Model Optimized via Two-Stage Rollout and Unified Criteria
Xinyu Hu, Yancheng He, Weixun Wang +6
Automatic evaluation is crucial yet challenging for open-ended natural language generation, especially when rule-based metrics are infeasible. Compared with traditional methods, th…