SWE-Prime introduces a two‑stage, multi‑granularity supervised fine‑tuning (SFT) data selection process for large language models tackling real‑world software problems. The first stage filters entire trajectories based on process quality, result quality, and representativeness, while the second stage evaluates semantic segments for contribution, learnability, and risk, keeping all segments in context but only penalizing selected ones during training. Experiments on SWE‑Bench Pro and Verified demonstrate that training on just 10% of trajectories chosen by SWE‑Prime outperforms full‑dataset training, achieving up to 12.2% and 24.2% relative gains.
MetaCurate is a training‑free framework that automatically evolves curation filters for agentic coding data by using pairwise large language model judgments to refine its rules, rubric, and aggregation policy. Starting from no initial filter, it discovers a strategy that scores trajectories independently, requiring pairwise comparisons only during the search phase. Experiments on an uncurated trajectory pool and the Nemotron v3 SWE dataset show that MetaCurate achieves the highest mean SWE‑Verified score compared to random selection, hand‑authored filtering, and a training‑loop baseline, while incurring lower search costs.
By Donghoon Kim, Seunghyuk Oh, Kevin Galim, Minseo Kim, Minjae Lee, Younghyun Kim, Wooseok Seo, Hyung Il Koo, Wonjun Kang
arXiv:2609.05837v1 Announce Type: new
Abstract: LLM-based agents are increasingly deployed in real-world applications through tool-use APIs, yet training them for specific environments remains fundam...
By Zhiyi Lyu, Yewen Li, Longtao Zheng, Shengtian Yang, Lang Feng, Lei Feng, Peng Jiang, Kun Gai, Qingpeng Cai, Bo An
The paper introduces TQS, a Trajectory Quality Score that evaluates coding-agent interaction trajectories based on structural signals such as information focus, intent–action alignment, and policy dynamics, rather than just final outcomes. TQS operates in linear time without requiring an LLM judge and is applied to 155,144 successful trajectories from a public corpus. Fine‑tuning agents on trajectories ranked highest by TQS consistently outperforms models trained on random or low‑ranked subsets on the SWE‑Bench‑Verified benchmark, demonstrating the value of process‑level quality filtering.
By Sida Li, Yuanxin Liu, Shicheng Li, Linli Yao, Xu Sun, Haoli Bai, Xiaohui Li
arXiv:2609.08318v1 Announce Type: cross
Abstract: The transition from human-centric assistance to Autonomous Software Engineering (ASE) agents has enabled the resolution of complex real-world SE task...
By Zhengran Zeng, Yixin Li, Rui Xie, Wei Ye, Shikun Zhang
arXiv:2609.18417v1 Announce Type: new
Abstract: Multi-turn agent trajectories often contain redundant rounds (failed tool calls, parallel sub-queries, verification-only steps) that inflate both train...
By Zhuo Chen, Zhen Zhang, Xinyu Wang, Kewei Tu
SWE-Critix is a post‑training pipeline that creates generalist verifiers for coding‑agent trajectories. It uses supervised fine‑tuning followed by reinforcement learning to train 30B models on CodeForge‑Preview data, enabling the verifier to predict completion labels and generate natural‑language explanations. The approach raises classification accuracy from 57% to 78% on a held‑out test set.
By Jiajun Hu, Chun Yong Chong
arXiv:2610.33568v1 Announce Type: new
Abstract: Successful tool-use trajectories are not necessarily useful supervision. Even successful interactions may underrepresent multi-step dependencies or con...
By Yihao Hu, Zhihao Wen, Yitong Lin, Siyuan Lu, Pan Wang, Wei Wu, Jiang-Ming Yang, Tao Lin
UniToolCall introduces a unified framework for tool-use in large language model agents, standardizing toolset construction, dataset generation, and evaluation. The framework aggregates over 22,000 tools and creates a hybrid training corpus of more than 390,000 instances by combining ten public datasets with synthetically generated, structurally controlled trajectories. It models diverse interaction patterns—single‑hop vs. multi‑hop, single‑turn vs. multi‑turn, serial vs. parallel execution—and adds an Anchor Linkage mechanism to enforce cross‑turn dependencies, while converting seven public benchmarks into a common Query–Action–Observation–Answer format for fine‑grained evaluation.
By Yijuan Liang, Xinghao Chen, Yifan Ge, Ziyi Wu, Hao Wu, Changyu Zeng, Wei Xing, Xiaoyu Shen
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
PlanPO introduces a group planning-aware policy optimization method for multi-turn agentic large language models, addressing the issue of advantage collapse caused by treating all successful trajectories equally. By incorporating coarse-to-fine advantage signals that reflect differences in trajectory and turn lengths, PlanPO encourages agents to learn generalizable planning and generation behaviors. Experiments show a 27.2% average improvement over GRPO on benchmarks such as ALFWorld, WebShop, and SciWorld, with minimal extra training cost.
By Dayang Liang, Liyuan He, Xuan Feng, Shuxin Li, Bo An, Yunlong Liu
arXiv:2602.16429v2 Announce Type: replace
Abstract: Agentic systems, AI architectures that autonomously execute multi-step workflows to achieve complex goals, are often built using repeated large lan...
By Ido Levy, Segev Shlomov, Avi Yaeli, Nir Mashkif, Asaf Adi, Yinon Goldshtein, Eilam Shapira, Leshem Choshen