From AutoResearch to Continual AutoResearch: Do Agents Learn Across Tasks?
- Published
- Source
- arXiv
- Paper number
- 1411
- Field
- Research Agents
- arXiv ID
- 2611.05108
Key points
- The authors propose CONTINUAL AUTORESEARCH, which pairs persistent and reset executions of the same agent under matched task streams, data, and time budget.
- The evaluation covers nine task families spanning synthetic tasks, symbolic regression, time-series forecasting, and tabular AutoML.
- The authors evaluated five frontier CLI agents, and persistent state improved fixed-budget research efficiency for all five.
- Time-convergence gains were +21.3 points for Kimi K3, +20.9 for Sonnet 4.6, +7.5 for Opus 5, +6.1 for GPT-5.6-Sol, and +3.0 for GPT-5.4.
- Sonnet 4.6 was the only agent with a substantial mean final outcome gain of +7.6 points, while the others were near zero or slightly negative.
- In the memory ablation, matched insight cards changed initialization and search but did not consistently improve final performance.
Paper links
External research summaries. These are not HDATF publications or measured product results.