13 papers
Navalia: Advancing Cowork Agents through Verifiable Task Synthesis and Long-Horizon Post-Training
Hengqi Liu, Ziyuan Liu, Yang Qin +7
Large language model (LLM) agents are evolving from conversational assistants into Cowork Agents capable of autonomous planning, tool use, and complex task execution. Financial ana…
Vela: User Behavior as Supervision Sparks LLM Intelligence
Yang Xiao, Yusong Sun, Haoyi Wu +17
Large language models are increasingly deployed in applications that act on users' long-term behavior histories, from personal assistants to recommendation and advertising systems.…
AdaptEvo: Learning under Imperfect Supervision through Confidence-Adaptive Training with Knowledge Evolution
AllSpark Team, Shijun Wan, Jiancong Xie +8
Rule-governed contextual decision tasks require models to apply specified rules to case-specific context and evidence. Written rules can leave gaps in decision guidance and process…
ROSS: Relearning from Self-Generated Rollouts through Selective Supervision
Zhiwei Zhang, Huayu Deng, Fei Zhao +4
Large language model post-training generates self-generated rollouts through reinforcement learning and on-policy distillation, yet this experience is often treated as stale once t…
Skill2Env: Capability-Oriented Environment Synthesis from Skills for General Agents
Weiyi Xu, Xiaowen Yang, Wen Da +7
Executable environments are critical for post-training agents on tasks that require tool use and multi-step interaction, but constructing executable tasks together with their envir…
CompoWorld: Compositional Environment Scaling for General Agents
Xiao-Wen Yang, Weiyi Xu, Wen Da +9
Automatically generated environments provide a scalable source of interaction data for training general agents. However, existing approaches mainly generate tasks within a single e…
Learning to Learn from Context: Synthetic Training from Perturbed Public Documents
Haoyi Wu, Yang Xiao, Yusong Sun +4
Real-world tasks often require large language models (LLMs) to learn from complex task-specific context rather than pretrained parametric knowledge. This capability remains a weakn…
PACT: From Credit Assignment to Critic Alignment
Jiayan Fu, Hang Xu, Yong Zhang +4
Reinforcement learning has become a central component of large language model (LLM) post-training, yet token-level credit lacks a generally accepted mathematical definition, leavin…
NoteVQA: Benchmarking VLMs on Real-Life Questions from Human Communities
Haonan Jiang, Guojian Zhan, Jiancong Xie +6
Vision-language models (VLMs) increasingly power consumer-facing AI search, yet evaluating them on the diversity of everyday visual questions remains challenging. Existing benchmar…
Iris: Climbing to the Search Frontier
Ziyuan Liu, Hengqi Liu, Zichuan Wang +8
We present Iris-mini and Iris-pro, two search agents trained at the 35B-A3B and 397B-A17B scales, together with the data pipeline and training recipe behind them. Tasks are reverse…
Verify Before You Distill: Prompt-Level Teacher Gating for On-Policy Distillation
Zhiwei Zhang, Zechen Sun, Fei Zhao +6
On-policy distillation (OPD) accelerates post-training by providing dense token-level supervision from a frozen teacher on the student's own rollouts. Vanilla OPD applies this supe…
PILOT in the Loop: Live Self-Improvement for Long-Horizon Agents
Yang Xiao, Yusong Sun, Haoyi Wu +7
Long-horizon agent runs generate experience that can improve both the current run and future work. Most self-improvement methods process this experience only after execution ends,…
D-MOPD: Dynamic Domain ScheDuling for Efficient Multi-Teacher Distillation
Zechen Sun, Zhiwei Zhang, Fei Zhao +7
Multi-teacher on-policy distillation (MOPD) distills several domain-expert teachers into a single student by minimizing per-domain reverse-KL divergence on the student's own rollou…