4 papers
Which Successes Make Good References? Action-Sequence Alignment for Multi-Turn On-Policy Distillation
Mufei Li, Nikki Lijing Kuang, Zhuoer Wang +9
On-policy distillation (OPD) reduces train-inference distribution mismatch by supervising a student on its own trajectories. For multi-turn tool-use agents, conditioning the teache…
UNBIND: UNlearning By INference-time Directional Steering for Code LLMs
Zhengyang Shan, Jiayun Xin, Yanjun Lin +7
Code large language models acquire programming capabilities from large code corpora, but can also memorize implementations that later require removal. Code unlearning is needed to…
Optimizing What Policies Learn From: Recoverability-aware Rollout Intervention Learning
Zheyuan Zhang, Manqing Mao, Hong Wang +8
Critic-free group-based reinforcement learning has become a scalable approach for post-training large language models. However, most existing methods allocate the same number of ro…
ReacTOD: Bounded Neuro-Symbolic Agentic NLU for Zero-Shot Dialogue State Tracking
Yanjun Lin, Zimo Xiao, Kartik Natarajan +7
Task-oriented dialogue systems -- handling transactions, reservations, and service requests -- require predictable behavior, yet the moderately-sized LLMs needed for practical late…