Experience Scaling: Post-Deployment Evolution For Large Language Models
This addresses the problem of diminishing returns in LLM development for AI researchers and practitioners, offering a scalable path for continued progress, though it appears incremental as it builds on existing scaling paradigms.
The paper tackles the saturation of scaling model size, training data, and compute power for large language models (LLMs) by proposing experience scaling, a framework for continuous post-deployment evolution through autonomous interaction and collaborative sharing, which improves accuracy and sustains performance in simulated real-world scenarios.
Scaling model size, training data, and compute power have driven advances in large language models (LLMs), but these approaches are reaching saturation as human-generated text is exhausted and further gains diminish. We propose experience scaling, a framework for continuous post-deployment evolution for LLMs through autonomous interaction with the environment and collaborative sharing of accumulated experience. The framework captures raw interactions, distills them into compact, reusable knowledge, and periodically refines stored content to preserve relevance and efficiency. We validate the framework in simulated real-world scenarios involving generalization to previously unseen but related tasks, repetitive queries, and over-saturated knowledge stores. Across all settings, experience scaling improves accuracy, sustains performance over time, and maintains gains when applied to novel situations. These results demonstrate that structured post-deployment learning can extend LLM capabilities beyond the limits of static human-generated data, offering a scalable path for continued intelligence progress.