arXiv:2608. 08466v1 Announce Type: new Abstract: Modern LLM agents are often improved by modifying prompts, tools, or workflows manually, while the executable scaffold surrounding the model---the \emph{harness}---is typically treated as a fixed artifact after deployment.
By Tailin Zhou
The paper introduces Latent Harness, a method that compiles external agent behaviors into a small rank‑8 activation shift applied to every decoder block of a frozen language‑model backbone, affecting only 0.03‑0.05% of parameters. This compiled shift eliminates the need for on‑call harness text, offers runtime controls for injection strength, phase, and mixing, and recovers most of the performance gains of the original harness across multiple settings. The approach also enables zero‑shot transfer across tasks, composability in shift space, and improves budget‑matched harness search, outperforming LoRA with far fewer parameters.
By Ran Li, Shimin Di, Zeyu Ren, Ling Yue, Jiachuan Wang, Lei Chen
The paper investigates whether a language model’s inherent task‑solving ability predicts its ability to self‑evolve its external harness—comprising prompts, skills, memories, and tools—by updating these components based on execution evidence. Two capabilities are examined: harness‑updating (producing useful updates) and harness‑benefit (gaining from those updates). Findings show that harness‑updating performance is largely independent of base capability, while harness‑benefit peaks at mid‑tier models and declines for both weaker and stronger models, suggesting that resources should focus on improving the agent’s core task‑solving skills rather than its evolution mechanism.
By Minhua Lin, Juncheng Wu, Zijun Wang, Zhan Shi, Yisi Sang, Bing He, Zewen Liu, Tianxin Wei, Zongyu Wu, Zhiwei Zhang, Dakuo Wang, Xiang Zhang, Benoit Dumoulin, Cihang Xie, Yuyin Zhou, Suhang Wang, Hanqing Lu
Agentic AI development today runs on token maxing: buying capability with tokens -- longer reasoning traces, more turns, wider tool payloads, bigger replayed contexts -- so tokens per task grow faster than task value. Falling per-token prices mask the pattern; total spend rises anyway.
The paper introduces Harness-Delta Attribution (HDA), a protocol that breaks down the performance gains from harness evolution into overfitting, test-time scaling, and residual generalizable improvement. Applying HDA to 11 executor models and three harness-evolution frameworks across various tasks, the study finds that over half of the search-set gains are due to overfitting or scaling, with limited transfer to held-out tasks. The results highlight the need for attribution-aware evaluation in future harness evolution research.
By Wenxuan Ding, Qin Lu, Changlong Yu, Sha Li, Shuowei Jin, Xin Liu, Zhenyu Shi, Greg Durrett
arXiv:2608. 09629v1 Announce Type: new Abstract: Self-evolving agents are usually built around prescribed optimization pipelines: the framework decides how to gather evidence, revise a persistent artifact, select candidates, and stop.
By Hui Xue, Fan Yang
arXiv:2608. 07545v1 Announce Type: cross Abstract: An LLM agent's capability depends not only on model weights but on its harness: prompts, tools, skills, and control flow.
By Yifan Zhang, Yutong Dai, Juntao Tan, Luyu Yang, Rishi Mullur, Thai Hoang, Zhiyuan Hu, James Zhu, Phil Mui, Silvio Savarese, Ran Xu, Zeyuan Chen
arXiv:2609.09134v1 Announce Type: new
Abstract: Agent harnesses (the system prompt, tool set, execution hooks, and context-management scaffolding around a model) are a critical determinant of agentic...
By Zhou Yu, Bin Bi, Shiva Kumar Pentyala, Shubham Mehrotra, Sougata Chaudhuri, Shilpa Bhagavath, Zeyuan Chen, Ran Xu, Phil Mui, James Zhu, Sitaram Asur
arXiv:2607. 06906v1 Announce Type: new Abstract: Agentic AI development today runs on token maxing: buying capability with tokens -- longer reasoning traces, more turns, wider tool payloads, bigger replayed contexts -- so tokens per task grow faster than task value.
By Muayad Sayed Ali, Aliaksandra Novik, Anji Boddupally, Artem Yavorskyi, Chris Nickerson, Daniel Rica, Emily DuGranrut, Felix Leung, Garrett Prince, Grace Barnett, Heath Robinson, Hosain Al Ahmad, Jesse Resnick, Juan Carlos Farah, Jyothi Swaroop Meruga, Leonid Kuznetsov, Luke Gorham, Marie Schmoll, Michael Paciullo, Saumya Das, Sharath Sheripally, Tommy Griscom, Mykyta Osadchyi, Neha Mantri, Nick Westrum, Olivia Benowitz, Parikshith Kulkarni, Radik Chernyshov, Rakshith Vasudev, Rohith Nadimpally, Vikas Gangadevi, Waseem AlShikh
The paper investigates when to evolve a harness versus training model weights for long‑horizon LLM agents. By labeling failed trajectories with the first signal that fires, the authors separate process failures from content failures, showing that harness evolution fixes the former while weight training addresses the latter. Experiments on DeepPlanning and WebArena‑Lite demonstrate significant performance gains, with LoRA adapters trained on evolved harness trajectories further improving held‑out scores and reducing content failures.
By Yuan Tian, Bing Hu, Hao Wang, Binghang Lu, Fang Wu
The paper introduces Regularized Recursive Self-Improvement of Agent Harnesses (RRSI), a method that applies regularization principles to the iterative editing of an LLM agent’s harness—prompts, control flow, tooling, memory, and context management. RRSI limits the number of edits per candidate, encourages novel trajectories, and uses a critic and pruner to filter out benchmark‑specific or ineffective changes, thereby favoring reusable agent mechanisms. Experiments on eight benchmarks show RRSI improves performance by up to 14.1 points on the training split and 4.7 points on out‑of‑distribution tests, while reducing policy token usage by 30% compared to unregularized evolution.
By Peng Xia, Rujun Han, Zifeng Wang, Yanfei Chen, Yufan Zhang, Yoonho Lee, Chengsong Huang, Han Yu, Zhongying CuiZhu, Yifei Ming, Huaxiu Yao, Burak Gokturk, Tomas Pfister, Chen-Yu Lee
The paper introduces a framework for self‑evolving harnesses in which a frozen language‑model agent first solves tasks and then, as a proposer, edits its own harness based on complete run records. The method trains on diverse benchmarks, separating training and held‑out tasks, and evaluates on five unseen out‑of‑distribution benchmarks. Experiments show that harnesses evolved by GLM‑5.3 and GPT‑5.6 Sol achieve performance comparable to or exceeding Claude Code and Codex, with further improvements after continued evolution on Claw‑Eval.
By Qiankai Xu