Do Agent Optimizers Compound? A Continual-Learning Evaluation on Terminal-Bench

arXiv:2604.05013v1 Announce Type: cross Abstract: Current LLM coding agents are predominantly trained on composite benchmarks (e.g., bug fixing), which often leads to

arXiv:2610.00025v1 Announce Type: new Abstract: Agent harnesses increasingly want to run small language models (SLMs) on the microtasks around a frontier large language model

arXiv:2605.11556v1 Announce Type: new Abstract: Solving complex long-horizon tasks requires strong planning and reasoning capabilities. Although datasets with explicit

arXiv:2603.18620v1 Announce Type: cross Abstract: We introduce Learning to Self-Evolve (LSE), a reinforcement learning framework that trains large language models (LLMs) to

arXiv:2605.08354v1 Announce Type: new Abstract: Aligning multimodal generative models with human preferences demands reward signals that respect the compositional,

arXiv:2607.22157v1 Announce Type: new Abstract: AI agents encounter learning opportunities in every episode they run, and discard nearly all of them: the underlying models are

arXiv:2604.22748v1 Announce Type: new Abstract: As AI systems move from generating text to accomplishing goals through sustained interaction, the ability to model environment

arXiv:2604.22436v1 Announce Type: new Abstract: The rapid growth of AI agent ecosystems is transforming how complex tasks are delegated and executed, creating a new challenge of