HS
Huatong Song
cs.CLcs.AIcs.LGcs.SEcs.IR
On Valency
published · living versionsW_eeq3w9uy·v1 · currentpublished
Multiscale Policy Optimization for Long-Horizon LLM Agents
with Tao Ren, Hui Yang, Jia Deng, Tingyang Chen +14
1 version
W_j4fhuxv6·v2published
Preprints & journals
13 papers in the corpus · 2025–2026IQuest-Q1: Advancing Foundation Capabilities for Agentic CLI Systems2610.30949v1 · A. Jian Yang, An Huang, Bo Dan et al.2026 · 0 citationsarXiv
Harness Scaling for Generalizable Agentic Reinforcement Learning2610.29458v2 · Fei Bai, Huatong Song, Chuan Hao et al.2026 · 0 citationsarXivon Valency
RealCLI: Toward a Unified Evaluation of Human-Agent Collaboration for CLI Agents2610.22507v1 · Yunqi Liu, Peiyu Wang, Bingli Wang et al.2026 · 0 citationsarXiv
Agent-Editing World Model: Rethinking World Modeling for LLM Agents2609.28416v1 · Shuang Sun, Guoxin Chen, Fanzhe Meng et al.2026 · 0 citationsarXiv
ClawGym II: Exploring Black-Box RL on Agent Harness2608.16798v1 · Huatong Song, Fei Bai, Ming Yang et al.2026 · 0 citationsarXiv
Evo-Bench: Can Language Models Improve Agent Harness?2608.09096v2 · Lisheng Huang, Chen Yang, Hao Zhou et al.2026 · 0 citationsarXiv
Nanbeige4.2-3B: Unlocking Agentic Capabilities in a Compact Model2607.22083v2 · Nanbeige Lab: Chen Yang, Chengrui Huang, Fufeng Lan et al.2026 · 0 citationsarXiv
ClawGym: A Scalable Framework for Building Effective Claw Agents2604.26904v3 · Fei Bai, Huatong Song, Shuang Sun et al.2026 · 0 citationsarXiv
BeyondSWE: Can Current Code Agent Survive Beyond Single-Repo Bug Fixing?2603.03194v2 · Guoxin Chen, Fanzhe Meng, Jiale Zhao et al.2026 · 0 citationsarXiv
SWE-Master: Unleashing the Potential of Software Engineering Agents via Post-Training2602.03411v2 · Huatong Song, Lisheng Huang, Shuang Sun et al.2026 · 0 citationsarXiv
SWE-World: Building Software Engineering Agents in Docker-Free Environments2602.03419v1 · Shuang Sun, Huatong Song, Lisheng Huang et al.2026 · 0 citationsarXiv
Computer Environments Elicit General Agentic Intelligence in LLMs2601.16206v3 · Daixuan Cheng, Shaohan Huang, Yuxian Gu et al.2026 · 0 citationsarXiv
R1-Searcher: Incentivizing the Search Capability in LLMs via Reinforcement Learning2503.05592v2 · Huatong Song, Jinhao Jiang, Yingqian Min et al.2025 · 7 citationsarXiv
Career total: 14 works. 13 are in this corpus.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.