arXiv AI October 2, 2026 By Beining Wu, Zihao Ding, Jun Huang

Not All Experience Belongs in the Weights: Component Routing for Self-Improving GUI Agents

The paper introduces component routing for self‑improving GUI agents, separating experience into locators, procedures, state facts, and lessons, and directing each to either the model weights or the prompt context. Experiments across three backbone families, two environments, and multiple seeds show that routing improves performance over whole‑trajectory baselines, with a rule based on recurrence and state‑conditionality accurately predicting the optimal destination. The study also analyzes how training dynamics and producer‑consumer differences affect the value of each destination, revealing that readout decreases for frequently recurring items when written to weights, while context gains grow with the information gap and weight gains shrink with the policy gap.

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 AI.

arXiv AI

Aug 18

Routing Divergence Is Not Evidence of Behavioral Influence in Same-Weight MoE Self-Distillation

arXiv:2608. 15787v1 Announce Type: cross Abstract: Two Mixture-of-Experts (MoE) forward passes can share every weight yet route the same token through different experts.

By Cedric Caruzzo, Donggeun Yoo, Tae Soo Kim

efficiency

arXiv Machine Learning

Aug 6

EvolveNet: Collaborative Harness Evolution for Agent Self-Improvement

arXiv:2608. 04968v1 Announce Type: new Abstract: The capabilities of an LLM agent depend not only on its model but on the harness: the executable program that constructs context, invokes tools, verifies results, and recovers from failure.

By Jun Nie, Yonggang Zhang, Qianshu Cai, Yiu-ming Cheung, Xinmei Tian, Bo Han

llms agents data-engineering

arXiv Computation and Language

Sep 22

Harness-Zero: Harness Distillation via Agent-as-Harness

arXiv:2609.24974v1 Announce Type: cross Abstract: Agent harnesses, the external systems that mediate model-environment interaction, can substantially improve agent performance, but their gains remain...

By Haoran Ye, Yuxing Lu, Haonan Dong, Zhaochen Su, Guojie Song

llms agents fine-tuning efficiency

arXiv AI

Sep 15

The Router Within: Eliciting Native Skill Routing from a Frozen LLM

arXiv:2609.15982v1 Announce Type: cross Abstract: Skills extend an LLM agent beyond its parametric knowledge, and the gain they promise rests on picking the right one. Deployed harnesses route by pre...

By Ruishuo Chen, Xun Wang, Yu Chen, Zhuoran Li, Longbo Huang

llms agents benchmarks

arXiv Computation and Language

Aug 27

AEL: Evolving Agent Harness in Open-Ended Environments

The paper introduces Agent Evolving Learning (AEL), a two‑timescale framework that dynamically evolves an LLM agent’s memory‑retrieval harness in open‑ended environments. A fast Thompson‑Sampling bandit selects among retrieval policies each episode, while a slower LLM reflection diagnoses performance drops and injects new policies when the current set plateaus. AEL outperforms ten self‑improving and non‑LLM baselines on a sequential portfolio benchmark, boosting Sharpe ratio by 27% and achieving significant accuracy gains on a support‑ticket routing stream.

By Wujiang Xu, Jiaojiao Han, Minghao Guo, Kai Mei, Xi Zhu, Han Zhang, Dimitris N. Metaxas

llms agents reinforcement-learning benchmarks

Hugging Face Trending Papers

Jul 30

Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale

Computer-use agents learn from what their actions change, so training one needs applications it can act on, break and reset. The applications that matter most are login-gated and stateful, so synthetic environments stand in for them.

agents benchmarks