9 papers
DASH-OPD: Discrepancy-Aware Switching with Hysteresis for On-Policy Distillation
Yuchen Xia, Qianguo Sun, Chao Song +3
While on-policy distillation (OPD) reduces exposure bias by training student language models on their own rollouts, early student errors in long-horizon agentic scenarios can lead…
Escaping the Self-Confirmation Trap: An Execute-Distill-Verify Paradigm for Agentic Experience Learning
Shiding Zhu, Yudi Qi, Yajie Wang +6
Experience-driven self-evolution is critical for large language model (LLM) agents to improve through open-world interaction. However, existing experience learning methods mostly r…
Demystifying Reinforcement Learning for Long-Horizon Tool-Using Agents: A Comprehensive Recipe
Xixi Wu, Qianguo Sun, Ruiyang Zhang +4
Reinforcement Learning (RL) is essential for evolving Large Language Models (LLMs) into autonomous agents capable of long-horizon planning, yet a practical recipe for scaling RL in…
VSearcher: Long-Horizon Multimodal Search Agent via Reinforcement Learning
Ruiyang Zhang, Qianguo Sun, Chao Song +2
Large models are increasingly becoming autonomous agents that interact with real-world environments and use external tools to augment their static capabilities. However, most recen…
ACPO: Adaptive Curriculum Policy Optimization for Aligning Vision-Language Models in Complex Reasoning
Yunhao Wang, Ziting Li, Shuai Chen +6
Aligning large-scale vision-language models (VLMs) for complex reasoning via reinforcement learning is often hampered by the limitations of existing policy optimization algorithms,…
SAFETY-J: Evaluating Safety with Critique
Yixiu Liu, Yuxiang Zheng, Shijie Xia +4
The deployment of Large Language Models (LLMs) in content generation raises significant safety concerns, particularly regarding the transparency and interpretability of content eva…
A Framework to Implement 1+N Multi-task Fine-tuning Pattern in LLMs Using the CGC-LORA Algorithm
Chao Song, Zhihao Ye, Qiqiang Lin +2
With the productive evolution of large language models (LLMs) in the field of natural language processing (NLP), tons of effort has been made to effectively fine-tune common pre-tr…
AdaptSSR: Pre-training User Model with Augmentation-Adaptive Self-Supervised Ranking
Yang Yu, Qi Liu, Kai Zhang +7
User modeling, which aims to capture users' characteristics or interests, heavily relies on task-specific labeled data and suffers from the data sparsity issue. Several recent stud…
PASTO: Strategic Parameter Optimization in Recommendation Systems -- Probabilistic is Better than Deterministic
Weicong Ding, Hanlin Tang, Jingshuo Feng +17
Real-world recommendation systems often consist of two phases. In the first phase, multiple predictive models produce the probability of different immediate user actions. In the se…