The paper introduces Cast‑R1, an 8‑billion‑parameter agentic forecasting model that learns to orchestrate analytical and prediction tools through trajectory distillation and multi‑turn reinforcement learning with structured memory. The training pipeline first generates multi‑turn interaction trajectories using a teacher LLM, then fine‑tunes the agent with supervised learning and group relative policy optimization, guided by rewards for accuracy and temporal structure. Experiments across several datasets show that Cast‑R1 effectively equips smaller language models with tool‑augmented, multi‑turn forecasting capabilities within structured workflows.
By Xiaoyu Tao, Mingyue Cheng, Yaguo Liu, Xingpeng Gao, Qingsong Wen, Shijin Wang, Enhong chen
The paper introduces an agentic forecasting environment built on 2,100+ resolved Polymarket questions, where a language model (Qwen3.5-35B-A3B) learns to gather evidence during rollout via web search, page reading, and financial time series, all filtered to avoid post‑cutoff leaks. Training with single‑epoch GRPO and a Brier‑score reward improves calibration by 30‑40% and reduces search attempts, while the trained policy outperforms four frontier models in evidence‑based forecasting, achieving lower soft‑Brier scores at roughly 5% of the inference cost. The authors release the environment, dataset, and per‑rollout records as a reusable harness for temporal forecasting agents.
By Yusuf Afifi, Artur Kiulian, Anton Polishko, Mykola Khandoga, Hamudi Naanaa, Alina Krasnobrizha
The paper introduces EXPO, an algorithm that automatically optimizes the meta-prompt—specifically the task description and meta-instruction—for large language model agents in sequential decision-making tasks such as Bayesian optimization and multi-armed bandits. Building on adversarial bandit techniques to handle non-stationary rewards, the authors extend EXPO to EXPO-ES, which also optimizes exemplars (historical interactions) within the meta-prompt. Experiments demonstrate that these methods significantly improve the performance of LLM-based agents in sequential decision-making scenarios.
By Mingze Kong, Zhiyong Wang, Yao Shu, Zhongxiang Dai
COBRA‑Skills is a new framework that treats skill optimization for large language model agents as a budgeted sequential problem over a dynamically evolving candidate set. It uses contextual‑bandit prioritization to focus evaluations on promising or informative candidates and refines the skill population based on execution feedback. In experiments across six agent benchmarks and three target models, COBRA‑Skills outperforms existing methods, cuts optimization cost by 55–58 % compared to SkillOpt, and requires only 50 unique optimization examples per benchmark.
By Pingchen Lu, Xiangyi Wang, Xiang Li, Jie Mao, Zikun Qu, Junfeng Luo, Yao Shu, Bryan Kian Hsiang Low, Zhongxiang Dai
CastFlow is a role‑specialized agentic forecasting framework that combines general‑purpose tool reasoning with specialized temporal forecasting. It uses an Experience Memory to guide a frozen LLM in acquiring evidence via a multi‑view toolkit, then a lightweight specialized LLM refines predictions through evidence‑grounded reasoning. The two‑stage training strategy—supervised fine‑tuning followed by reinforcement learning with verifiable rewards—optimizes the specialized model based on final forecasting outcomes, achieving superior performance on nine out of ten real‑world benchmarks.
By Bokai Pan, Mingyue Cheng, Xiaoyu Tao, Zhiding Liu, Shuo Yu, Yuchong Wu, Qi Liu, Defu Lian, Enhong Chen
arXiv:2606. 15197v1 Announce Type: cross Abstract: Optimization modeling is inherently hierarchical, requiring a precise sequence of symbolic commitments.
By Jiajun Li, Yu Ding, Shisi Guan, Ran Hou, Wanyuan Wang
The paper introduces CoPE, a method that trains language agents by adding multi‑step action targets to the policy backbone. By supervising each action forecasted from preceding histories, CoPE enables acting and forecasting to co‑evolve without extra inference cost. Across five benchmarks, CoPE outperforms existing RL baselines, achieving up to twice the sample efficiency on ScienceWorld and improving overall performance on realistic applications.
By Haofei Yu, Yixing Chen, Yining Zhao, Xuyan Ye, Yuxiang Liu, Zhigang Hua, Yan Xie, Shuang Yang, Jiaxuan You
arXiv:2607. 24779v1 Announce Type: new Abstract: Online advertising bidding systems typically deploy multiple offline-trained expert models (e.
By Ji Wu, Yunshan Peng, Wentao Bai, Yunke Bai, Wenzheng Shu, Jinan Pang, Yanxiang Zeng, Xialong Liu
The paper introduces Strategy Accumulation and Guided Execution (SAGE), a two-stage framework that makes automated fine-tuning of large language models cumulative. In the first stage, a multi-agent pipeline uses Monte Carlo Tree Search to explore training strategies while a Distillation Agent records task-specific insights and cross-task confidence scores into a structured repository. In the second stage, SAGE retrieves relevant experience from this repository to guide training on new tasks, achieving a 12.4‑percentage‑point improvement over a baseline pipeline without accumulated experience on nine unseen tasks.
By Haoran Zhao, Wei Du, Dingwen Yang, Jixuan Huang, Junlin Shang, Lingyong Fang, Ya Guo, Tao Gui, Qi Zhang, Xuanjing Huang
arXiv:2610.14913v1 Announce Type: new
Abstract: Reinforcement learning (RL) is essential for adaptive decision-making, e.g., robust Meta-RL and domain randomization (DR), and efficient post-training...
By Jiarong Wen, Qi Tao, Kaiyu Zhang, Yun Qu, Lizhou Cai, Heming Zou, Yuhang Jiang, Qi Wang
Agentick is a unified benchmark for sequential decision‑making agents that evaluates RL, LLM, VLM, hybrid, and human agents on 37 procedurally generated tasks across six capability categories, four difficulty levels, and five observation modalities via a single Gymnasium‑compatible interface. It includes a Coding API, oracle reference policies, pre‑built SFT datasets, a composable agent harness, and a live leaderboard. An evaluation of 27 configurations and over 90,000 episodes shows no single approach dominates, with GPT‑5 mini leading overall, PPO excelling in planning and multi‑agent tasks, and the reasoning harness boosting LLM performance by 3‑10×, while ASCII observations outperform natural language.
By Roger Creus Castanyer, Pablo Samuel Castro, Glen Berseth
The paper introduces a two‑stage framework for reflective prompt optimization that separates candidate generation from validation. In the construction stage, multiple parents are sampled and their training responses reused across parallel reflection branches, with each revision screened on a small minibatch to allow locally improved candidates to propagate without full validation. In the exploration stage, a fixed full‑validation budget is used to select the best prompt from a frozen candidate pool, enabling independent control of pool size and validation effort. Experiments on five benchmarks with Qwen3.5‑9B and Llama‑3.1‑8B‑Instruct show that retaining 100 candidates and performing 20 full validations can improve performance by up to 11.22 percentage points over the initial solver, and learned selection consistently outperforms the initial solver across all comparisons.
By Xiaofan Zhou, Lu Cheng