arXiv Computation and Language September 1, 2026 By Yuhao Wu, Jingyuan Zhang, Jiajun Shi, Yuxuan Zhang, Xinping Lei, Junting Zhou, Zexuan Wang, Yuchen Wu, Huan Zhou, Duo Wang, Yinzhu Piao, Yongchang Peng, Yunfeng Shi, Jin Chen, Zuo Wang, Jinkai Liu, Jiaheng Liu, Wenxuan Zhang, Shen Yan, Wenhao Huang, Ge Zhang

Aspire: Can Models Self-Evolve from Vague Goals?

Read the original on arXiv Computation and Language →

The paper introduces ASPIRE, a benchmark that challenges language model agents to self‑evolve from vague, natural‑language goals without explicit evaluation metrics. In ASPIRE, agents must interpret the goal, select data and update strategies, and decide when to evaluate, all while the downstream tasks remain hidden. Experiments show that while agents can complete training loops, weight‑level improvements are sparse and unstable, and the best evolved harness still falls short of a strong engineered baseline.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computation and Language.

Hugging Face Trending Papers

4d ago

Aspire: Can Models Self-Evolve from Vague Goals?

Many important forms of human learning begin with a vague goal, such as "become a better physicist" or "improve at research." Learners must interpret the goal, identify capability gaps, decide how to...

llms agents benchmarks

arXiv AI

Jun 11

GrowLoop: Self-Evolving Conversation Evaluation Seeded by Human

arXiv:2605. 28882v2 Announce Type: replace-cross Abstract: With the rapid advancement of large language models, evaluating human-likeness in open-ended conversation has become increasingly important.

By Yihang Lin, Yunze Gao, Zeyang Lin, Dongbo Li, Kun Peng, Yue Liu

llms agents benchmarks safety

arXiv AI

Jul 24

Self-Evolving Recommendation System: End-To-End Autonomous Model Optimization With LLM Agents

arXiv:2602. 10226v2 Announce Type: replace-cross Abstract: Optimizing large-scale machine learning systems, such as recommendation models for global video platforms, requires navigating a massive hyperparameter search space and, more critically, designing sophisticated optimizers, architectures, and reward functions to capture nuanced user behaviors.

By Haochen Wang, Yi Wu, Daryl Chang, Li Wei, Lukasz Heldt

llms agents efficiency

arXiv AI

Jul 22

Dr. Zero: Self-Evolving Search Agents without Training Data

arXiv:2601. 07055v2 Announce Type: replace Abstract: As high-quality data becomes increasingly difficult to obtain, self-evolution without curated training data has emerged as a promising paradigm.

By Zhenrui Yue, Kartikeya Upasani, Xianjun Yang, Suyu Ge, Shaoliang Nie, Yuning Mao, Zhe Liu, Dong Wang

llms agents nlp benchmarks

Hugging Face Trending Papers

Aug 10

Evo-Bench: Can Language Models Improve Agent Harness?

Large Language Models (LLMs) have driven rapid progress in autonomous agents, yet standard evaluations remain confined to static task solving. An emerging frontier is harness evolution---the agent's capacity to autonomously optimize its own operating harness.

llms agents benchmarks

arXiv Machine Learning

Jul 30

DREvo: Distilling Recalibrated Historical Experience for Harness Self-Evolution

arXiv:2607. 26722v1 Announce Type: cross Abstract: Harness plays a critical role in large language model agent performance, and building a high-performing harness requires substantial expert effort.

By Hanghui Guo, Weijie Shi, Zhangze Chen, Shengxiang Xu, Yishu Wang, Yimei Zhang, Wangze Ni, Jia Zhu, Shimin Di

llms agents efficiency benchmarks

One email a morning, machine-written

Related stories

Aspire: Can Models Self-Evolve from Vague Goals?

GrowLoop: Self-Evolving Conversation Evaluation Seeded by Human

Self-Evolving Recommendation System: End-To-End Autonomous Model Optimization With LLM Agents

Dr. Zero: Self-Evolving Search Agents without Training Data

Evo-Bench: Can Language Models Improve Agent Harness?

DREvo: Distilling Recalibrated Historical Experience for Harness Self-Evolution