9 papers
When Do Intrinsic Rewards Work for Code Reasoning? A Comprehensive Study
Xiaolong Jin, Xuandong Zhao, Wenbo Guo +2
Reinforcement learning with verifiable rewards (RLVR) has driven substantial progress in large language model reasoning, but relies on ground-truth supervision that is costly or in…
Skill-Guided Continuation Distillation for GUI Agents
Zhimin Fan, Hongwei Yu, Yeqing Shen +9
Improving GUI agents typically relies on behavior cloning on expert trajectories. However, as the current policy deviates from the expert policy, it inevitably encounters policy-in…
AndroidDaily: A Verifiable Benchmark for Mobile GUI Agents on Real-World Closed-Source Applications
Yifan Sui, Xin Huang, Hongbing Li +14
The rapid development of GUI foundation models and mobile GUI agents has spurred numerous evaluation benchmarks, yet most rely on simulated environments or open-source applications…
WIRE: Profiling Witnessed Within-Policy Instruction Collisions in LLM Agents
Lu Yan, Xuan Chen, Xiangyu Zhang
LLM agents are governed by long-lived prompt policies, where individually reasonable stand- ing rules can jointly govern the same pre- generation state. Existing instruction-follow…
Deep Researcher Agent: An Autonomous Framework for 24/7 Deep Learning Experimentation with Zero-Cost Monitoring
Xiangyue Zhang
We present \textbf{Deep Researcher Agent}, an open-source framework that enables large language model (LLM) agents to autonomously conduct deep learning experiments around the cloc…
When the Specification Emerges: Benchmarking Faithfulness Loss in Long-Horizon Coding Agents
Lu Yan, Xuan Chen, Xiangyu Zhang
Current coding-agent benchmarks usually pro- vide the full task specification upfront. Real research coding often does not: the intended system is progressively disclosed through i…
Who Tests the Testers? Systematic Enumeration and Coverage Audit of LLM Agent Tool Call Safety
Xuan Chen, Lu Yan, Ruqi Zhang +1
Large Language Model (LLM) agents increasingly act through external tools, making their safety contingent on tool-call workflows rather than text generation alone. While recent ben…
PaCoRe: Learning to Scale Test-Time Compute with Parallel Coordinated Reasoning
Jingcheng Hu, Yinmin Zhang, Shijie Shang +17
We introduce Parallel Coordinated Reasoning (PaCoRe), a training-and-inference framework designed to overcome a central limitation of contemporary language models: their inability…
StepFun-Prover Preview: Let's Think and Verify Step by Step
Shijie Shang, Ruosi Wan, Yue Peng +4
We present StepFun-Prover Preview, a large language model designed for formal theorem proving through tool-integrated reasoning. Using a reinforcement learning pipeline that incorp…