arXiv AI August 25, 2026 By Junda He, Jieke Shi, Zhou Yang, Mingfei Cheng, David Lo

Learning from the Test: Self-Referential Differential Testing for Deep RL Agents

The paper introduces Delta, a two‑phase framework for testing deep reinforcement learning agents. In the first phase, the agent under test is evaluated for catastrophic failures while collecting decision‑making data. The second phase trains a challenger agent from this data using offline RL; comparing the challenger’s rewards to the original agent reveals optimality bugs, and Delta successfully uncovered thousands of such issues across multiple environments.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

arXiv Machine Learning

Jul 28

Evaluating Fuzz Testing for Reinforcement Learning Agents

arXiv:2607. 24577v1 Announce Type: new Abstract: Reinforcement Learning (RL) agents are increasingly deployed in safety-critical domains such as robotics, autonomous driving, and drone control, where unexpected behaviors may lead to severe real-world consequences.

By Zhibin Kang, Hanmo You, Dong Wang, Haiming Zheng, Junjie Chen

agents reinforcement-learning robotics efficiency benchmarks safety

arXiv AI

Jul 13

Towards Shutdownable Agents: Generalizing Stochastic Choice in RL Agents and LLMs

arXiv:2604. 17502v4 Announce Type: replace Abstract: Misaligned artificial agents might resist shutdown.

By Carissa Cullen, Harry Garland, Alexander Roman, Louis Thomson, Christos Ziakas, Elliott Thornley

llms agents reinforcement-learning fine-tuning

arXiv Computation and Language

1d ago

Tool Verification for Test-Time Reinforcement Learning

arXiv:2603.02203v2 Announce Type: replace-cross Abstract: Test-time reinforcement learning (TTRL) has emerged as a promising paradigm for Recursive Self-Improving AI (RSI) by adapting Large Reasoning...

By Ruotong Liao, Nikolai R\"ohrich, Xiaohan Wang, Yuhui Zhang, Yasaman Samadzadeh, Volker Tresp, Serena Yeung-Levy

reinforcement-learning

arXiv AI

Aug 19

LEGO-RL: Harness-Native Reinforcement Learning for Coding Agents

LEGO-RL is a framework that connects native coding-agent harnesses with scalable policy‑gradient training without altering the harnesses’ internal flow. It achieves faithful optimization through in‑process LLM proxying, reliable execution via sandbox orchestration, and observable training with automated validation and a Live UI. Experiments show LEGO‑RL improves the Qwen3.5‑35B‑A3B model’s performance on three native harnesses while preserving high rollout‑training probability correlation.

By Yiming Du, Yuxin Jiang, Tao Yuan, Jianbo Dai, Shaowei Wang, Jierun Chen, Chaofan Tao, Xianzhi Yu, Lifeng Shang, Kam-Fai Wong, Xiaohui Li, Haoli Bai

llms agents reinforcement-learning safety

arXiv AI

Jun 2

LLM4Cov: Execution-Aware Agentic Learning for High-coverage Testbench Generation

arXiv:2602. 16953v3 Announce Type: replace Abstract: Execution-aware LLM agents offer a promising paradigm for learning from tool feedback, but such feedback can be expensive and slow to obtain, making online reinforcement learning (RL) less practical in certain scenarios.

By Hejia Zhang, Zhongming Yu, Chia-Tung Ho, Haoxing Ren, Brucek Khailany, Jishen Zhao

llms agents reinforcement-learning benchmarks

arXiv AI

Jul 31

Graph Is the Verifier: Agentic Reinforcement Learning for Interprocedural Vulnerability Detection

arXiv:2607. 26656v1 Announce Type: cross Abstract: Real-world vulnerabilities often span multiple functions, yet most learning-based detectors classify each function in isolation: on a sample of real CVEs, we find that 71.

By Yikun Li, Ting Zhang, Jiakun Liu, Jinfeng Jiang, Yuheng Yieh, Yixin Yang, Wen Bin Leow, Yide Yin, Yintong Huo, Eng Lieh Ouh, Lwin Khin Shar, David Lo

agents reinforcement-learning benchmarks

One email a morning, machine-written

Related stories

Evaluating Fuzz Testing for Reinforcement Learning Agents

Towards Shutdownable Agents: Generalizing Stochastic Choice in RL Agents and LLMs

Tool Verification for Test-Time Reinforcement Learning

LEGO-RL: Harness-Native Reinforcement Learning for Coding Agents

LLM4Cov: Execution-Aware Agentic Learning for High-coverage Testbench Generation

Graph Is the Verifier: Agentic Reinforcement Learning for Interprocedural Vulnerability Detection