7 papers · 1 filter
Beyond Token Scale: Chunk-Level Sparse Autoencoders for Reliable Semantic Feature Discovery
Xu Wang, Yifan Yang, TingHao YU +1
Sparse autoencoders (SAEs) expose features that help us understand and steer language models, but faithful reconstruction does not guarantee informative concepts. Token-level objec…
Environment Evolution for Terminal Agents
Zhiyuan Fan, Tinghao Yu, Yuanjun Cai +9
Scaling interactive and verifiable environments is critical for training terminal agents. As frontier models become more capable, environments synthesized from scratch become less…
DocPO: Advancing Document Policy Optimization via Tailored Step-Aware Rewards
Yunhao Wang, Binghong Wu, Zhenyu Huang +4
Reinforcement learning (RL) for document parsing often relies on reference-based rewards rooted in edit distance (e.g., tree edit distance), yet it remains hard to optimize in the…
Deep Research Pretraining via Predictive Navigation
Jiang Zhou, Zhiyuan Fan, Xing Wu +3
Deep research agents are often trained on expensive, environment-grounded tool-use trajectories that require repeated retrieval, document inspection, and report evaluation. We intr…
MORE: A Multilingual Document Parsing Benchmark and Evaluation
Long Xu, Binghong Wu, Tinghao Yu +6
Multilingual documents encapsulate rich regional cultures, scientific discoveries, and historical records. Parsing this content into structured, machine-readable formats is critica…
Toward Scalable Terminal Task Synthesis via Skill Graphs
Zhiyuan Fan, Tinghao Yu, Yuanjun Cai +8
Terminal agents have demonstrated strong potential for autonomous command-line execution, yet their training remains constrained by the scarcity of high-quality and diverse executi…
PolicyLong: Towards On-Policy Context Extension
Junlong Jia, Jiang Zhou, Ziyang Chen +5
Extending LLM context windows is hindered by scarce high-quality long-context data. Recent methods synthesize data with genuine long-range dependencies via information-theoretic ve…