arXiv AI September 11, 2026 By Rui Sun, Zhan Shi, Bing He

TRACE: Training Reasoning Agents for Causal Exploration with Synthesized Rewards

The paper introduces TRACE, a digital‑advertising diagnostic environment that uses simulated interventions to generate verifiable rewards for training reasoning agents. By injecting controlled interventions into a simulator, the hidden cause of anomalies becomes an oracle label, enabling agents to learn to identify root causes and affected segments through noisy, confounded evidence. Experiments show that reinforcement learning with these synthesized rewards outperforms large prompted baselines, achieving higher accuracy while using fewer tool calls.

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 AI

Sep 30

MLToolBench: Learning Tool-Augmented Agents for Machine Learning Development

arXiv:2609.36679v1 Announce Type: new Abstract: Machine learning engineering (MLE) agents have made substantial progress, but learning through ML experimentation remains costly in time and computatio...

By Xin Yu, Lizhu Zhang, Jiamu Bai, Yanhong Wu, Zellux Wang, Serena Li, Weiwei Li, Lingzhou Xue, Xiangjun Fan, Bo Peng

agents reinforcement-learning fine-tuning

arXiv AI

Oct 15

ACED-Bench: Evaluating How LLM Agents Acquire and Act on Causal Evidence

arXiv:2610.14381v1 Announce Type: new Abstract: Large language models are increasingly used as scientific assistants, yet we still lack reliable ways to evaluate whether they can investigate a system...

By Yiwei Chen, Amila Weerasinghe, Siddharth Biswal, Utkrisht Rajkumar, Olcay Boz, Taylor Berg-Kirkpatrick, Rose Yu

llms agents benchmarks

arXiv Machine Learning

Oct 15

The Invisible Leash: Why RLVR May or May Not Escape Its Origin

arXiv:2507.14843v5 Announce Type: replace Abstract: Recent advances highlight Reinforcement Learning with Verifiable Rewards (RLVR) as a promising approach to enhancing LLM capabilities. However, it...

By Fang Wu, Weihao Xuan, Danny Xing, Ximing Lu, Mingjie Liu, Yi Dong, Zaid Harchaoui, Yejin Choi

llms reinforcement-learning benchmarks

arXiv Computation and Language

Sep 17

LEEPS: Latent-Guided Explore-Exploit Prompt Sampling for Efficient RLVR in Large Language Models

arXiv:2607.28077v2 Announce Type: replace Abstract: Reinforcement learning with verifiable rewards (RLVR) improves the reasoning capabilities of large language models, but prompt groups with identica...

By Shuang Liang, Haoyang Zhou, Yifan Gong, Guowei Wang, Xiting Wang

llms reinforcement-learning benchmarks

arXiv AI

Oct 9

Proof-of-Use: Mitigating Tool-Call Hacking in Deep Research Agents

The paper introduces Proof-of-Use (PoU), a reinforcement learning framework designed to prevent tool‑call hacking in deep research agents. PoU enforces evidence grounding by requiring agents to cite normalized evidence identifiers and uses a multi‑objective reward system that includes progressive process rewards, an answer‑support alignment reward, and a curriculum‑style adaptive reward mixing. Experiments demonstrate that PoU effectively mitigates tool‑call hacking and promotes robust, adaptive tool usage even under domain and tool shifts.

By Shengjie Ma, Chenlong Deng, Jiaxin Mao, Jiadeng Huang, Teng Wang, Junjie Wu, Changwang Zhang, Jun wang

agents reinforcement-learning safety

arXiv AI

Oct 15

Verifiable Process Rewards for Agentic Reasoning

The paper introduces Verifiable Process Rewards (VPR), a framework that uses symbolic or algorithmic task-solving oracles as action-level verifiers to provide dense rewards for intermediate actions in agentic reasoning tasks. VPR is applied to Sokoban, Sudoku, and Minesweeper, and improves policy performance over outcome-reward baselines. The authors also demonstrate VPR’s effectiveness in a tool‑use benchmark, highlighting its reliance on reliable task oracles and the difficulty of extending it to less structured environments.

By Huining Yuan, Zelai Xu, Huaijie Wang, Xiangmin Yi, Jiaxuan Gao, Xiao-Ping Zhang, Yu Wang, Chao Yu, Yi Wu

llms agents reinforcement-learning

One email a morning, machine-written

Related stories

MLToolBench: Learning Tool-Augmented Agents for Machine Learning Development

ACED-Bench: Evaluating How LLM Agents Acquire and Act on Causal Evidence

The Invisible Leash: Why RLVR May or May Not Escape Its Origin

LEEPS: Latent-Guided Explore-Exploit Prompt Sampling for Efficient RLVR in Large Language Models

Proof-of-Use: Mitigating Tool-Call Hacking in Deep Research Agents

Verifiable Process Rewards for Agentic Reasoning