AgentEvolver: System-Wide Self-Evolution Through Task Execution
We present AgentEvolver, a system for developing capabilities during task execution while keeping the foundation model fixed.
Key points
- An agent can complete a task without improving how it works.
- Turning task experience into reusable capability requires connecting the changed component to its evaluation and subsequent use.
- We evaluate task outcomes on SWE-bench Pro Public and examine capability changes in six application cases.
- AgentEvolver provides a concrete basis for studying capability accumulation through execution; independent-task transfer and total development cost remain open questions.
Sources (1)
- [1]AgentEvolver: System-Wide Self-Evolution Through Task ExecutionarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 8, 09:55 AM
We present AgentEvolver, a system for developing capabilities during task execution while keeping the foundation model fixed.
An agent can complete a task without improving how it works.
Extractive summary: sentences quoted from the sources.
Before this
- Oct 8, 2026ConwayResearch/Underdog-Saluki-27B-1.0
- Oct 8, 2026Opera: A Verbal Critic Framework for Long-horizon Coding Agents
- Oct 7, 2026CoTrace: Data Recipes for Training Terminal Agents with Harness-Model Co-Evolution
- Oct 7, 2026Agent Lightning v1.0: A 3,500-Line Lightweight Agentic RL Framework for Training Agents with Real Harnesses
- Oct 7, 2026TestGRAD: Evolving Test Suites via Failure Pattern Momentum for SWE-Agent Ensemble
- Oct 7, 2026Code Understanding is a Bottleneck for Coding Agents