Beyond the Capability Boundary: Zeroth-Order Optimization for Self-Evolving LLM Agents

• URL: http://arxiv.org/abs/2608.09292v1
• Date: Mon, 10 Aug 2026 08:45:48 GMT
• Title: Beyond the Capability Boundary: Zeroth-Order Optimization for Self-Evolving LLM Agents
• Abstract summary: We propose a zeroth-order self-evolution framework that enables agents to learn beyond their capability boundary.<n>Specifically, we perturb LoRA parameters of LLMs, run the agent, compute the losses under the perturbed and original parameters, and use the loss difference to estimate gradients.
• Score: 39.47590544708589
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Self-evolving methods improve the capabilities of LLM agents by sampling trajectories from the underlying LLMs and learning from these trajectories. However, these methods struggle to learn beyond the inherent capability boundary of the agents, since the agents cannot sample correct trajectories on difficult examples for further improvements. In this paper, we propose a zeroth-order self-evolution framework that enables agents to learn beyond their capability boundary by perturbing LLM parameters to adapt to difficult examples without any trajectory annotations. Specifically, we perturb LoRA parameters of LLMs, run the agent, compute the losses under the perturbed and original parameters, and use the loss difference to estimate gradients and further update the LoRA parameters. We sample trajectories using the updated LLMs for supervised fine-tuning to break through the capability boundary of the agents, forming a closed self-evolution loop. We introduce a parallel perturbation inference mechanism and an adaptive lookup mechanism to reduce time consumption in zeroth-order optimization, with an answer perplexity loss that provides smooth and stable zeroth-order loss values. Experiments on multiple deep research benchmarks show that our method obtains substantially more successful trajectories and consistently outperforms strong baselines, especially on difficult examples. The code and released artifacts are available at https://github.com/hidk1911/ZOForLLMAgents.

Related papers

• Memory-Based Advantage Shaping for LLM-Guided Reinforcement Learning [18.215893951726166]
  In environments with sparse or delayed rewards, reinforcement learning incurs high sample complexity.<n>This limitation has motivated the use of large language models (LLMs) for subgoal discovery and trajectory guidance.<n>We address these challenges by constructing a memory graph that encodes subgoals and trajectories from both LLM guidance and the agent's own successful rollouts.
  arXiv  Detail & Related papers  (2026-02-20T01:44:35Z)
• MIRA: Memory-Integrated Reinforcement Learning Agent with Limited LLM Guidance [18.215893951726166]
  Large language models (LLMs) can provide subgoal decompositions, plausible trajectories, and abstract priors that facilitate early learning.<n>We propose MIRA (Memory-Integrated Reinforcement Learning Agent), which incorporates a structured, evolving memory graph to guide early training.
  arXiv  Detail & Related papers  (2026-02-20T01:43:30Z)
• Zero-Order Optimization for LLM Fine-Tuning via Learnable Direction Sampling [40.94400211806987]
  We propose a policy-driven ZO framework that treats the sampling distribution over perturbation directions as a learnable policy.<n>We show that learned sampling improves quality gradient information and relax the explicit dependence on $d$ in convergence bounds.<n>Our results suggest that adaptive direction sampling is a promising route to make ZO fine-tuning viable at scale.
  arXiv  Detail & Related papers  (2026-02-14T08:01:41Z)
• Stable Forgetting: Bounded Parameter-Efficient Unlearning in LLMs [30.089412595436585]
  We provide a theoretical framework that explains how ascent on the forget set destabilizes optimization in the feedforward layers of large language models (LLMs)<n>We propose Bounded Bounded Unlearning, a parameter-efficient approach that stabilizes fine-tuning by applying bounded functions to adapters.<n>Our method achieves substantial improvements in forgetting while preserving retention, establishing a novel theoretically grounded and practically scalable framework for unlearning in LLMs.
  arXiv  Detail & Related papers  (2025-09-29T01:30:15Z)
• Agentic Reinforced Policy Optimization [66.96989268893932]
  Large-scale reinforcement learning with verifiable rewards (RLVR) has demonstrated its effectiveness in harnessing the potential of large language models (LLMs) for single-turn reasoning tasks.<n>Current RL algorithms inadequately balance the models' intrinsic long-horizon reasoning capabilities and their proficiency in multi-turn tool interactions.<n>We propose Agentic Reinforced Policy Optimization (ARPO), a novel agentic RL algorithm tailored for training multi-turn LLM-based agents.
  arXiv  Detail & Related papers  (2025-07-26T07:53:11Z)
• Training LLM-Based Agents with Synthetic Self-Reflected Trajectories and Partial Masking [61.61356842567952]
  We propose STeP, a novel method for improving LLM-based agent training.<n>We synthesize self-reflected trajectories that include reflections and corrections of error steps.<n>Experiments demonstrate that our method improves agent performance across three representative tasks.
  arXiv  Detail & Related papers  (2025-05-26T14:11:12Z)
• Scaling Autonomous Agents via Automatic Reward Modeling And Planning [52.39395405893965]
  Large language models (LLMs) have demonstrated remarkable capabilities across a range of tasks.<n>However, they still struggle with problems requiring multi-step decision-making and environmental feedback.<n>We propose a framework that can automatically learn a reward model from the environment without human annotations.
  arXiv  Detail & Related papers  (2025-02-17T18:49:25Z)
• Mitigating Forgetting in LLM Fine-Tuning via Low-Perplexity Token Learning [65.23593936798662]
  We show that fine-tuning with LLM-generated data improves target task performance and reduces non-target task degradation.<n>This is the first work to provide an empirical explanation based on token perplexity reduction to mitigate catastrophic forgetting in LLMs after fine-tuning.
  arXiv  Detail & Related papers  (2025-01-24T08:18:56Z)
• Training Agents with Weakly Supervised Feedback from Large Language Models [19.216542820742607]
  This paper introduces a novel training method for LLM-based agents using weakly supervised signals from a critic LLM.<n>Our agents are trained in iterative manner, where they initially generate trajectories through environmental interaction.<n>Tests on the API-bank dataset show consistent improvement in our agents' capabilities and comparable performance to GPT-4.
  arXiv  Detail & Related papers  (2024-11-29T08:47:04Z)
• From Words to Actions: Unveiling the Theoretical Underpinnings of LLM-Driven Autonomous Systems [59.40480894948944]
  Large language model (LLM) empowered agents are able to solve decision-making problems in the physical world. Under this model, the LLM Planner navigates a partially observable Markov decision process (POMDP) by iteratively generating language-based subgoals via prompting. We prove that the pretrained LLM Planner effectively performs Bayesian aggregated imitation learning (BAIL) through in-context learning.
  arXiv  Detail & Related papers  (2024-05-30T09:42:54Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.