{"work":{"id":"3fb60853-ab99-4f3e-8f4c-27017c08f0be","openalex_id":null,"doi":null,"arxiv_id":"2603.16856","raw_key":null,"title":"Online Experiential Learning for Language Models","authors":null,"authors_text":"Tianzhu Ye, Li Dong, Qingxiu Dong, Xun Wu, Shaohan Huang, and Furu Wei","year":2026,"venue":"cs.CL","abstract":"The prevailing paradigm for improving large language models relies on offline training with human annotations or simulated environments, leaving the rich experience accumulated during real-world deployment entirely unexploited. We propose Online Experiential Learning (OEL), a framework that enables language models to continuously improve from their own deployment experience. OEL operates in two stages: first, transferable experiential knowledge is extracted and accumulated from interaction trajectories collected on the user side; second, this knowledge is consolidated into model parameters via on-policy context distillation, requiring no access to the user-side environment. The two stages are iterated to form an online learning loop, where the improved model collects higher-quality trajectories that yield richer experiential knowledge for subsequent rounds. We evaluate OEL on text-based game environments across multiple model scales and both thinking and non-thinking variants. OEL achieves consistent improvements over successive iterations, enhancing both task accuracy and token efficiency while preserving out-of-distribution performance. Our analysis further shows that extracted experiential knowledge is significantly more effective than raw trajectories, and that on-policy consistency between the knowledge source and the policy model is critical for effective learning.","external_url":"https://arxiv.org/abs/2603.16856","cited_by_count":null,"metadata_source":"pith","metadata_fetched_at":"2026-07-07T12:33:45.190220+00:00","pith_arxiv_id":"2603.16856","created_at":"2026-05-11T04:05:55.306332+00:00","updated_at":"2026-07-07T12:33:45.190220+00:00","title_quality_ok":true,"display_title":"Online experiential learning for language models","render_title":"Online experiential learning for language models"},"hub":{"state":{"work_id":"3fb60853-ab99-4f3e-8f4c-27017c08f0be","tier":"hub","tier_reason":"10+ Pith inbound or 1,000+ external citations","pith_inbound_count":16,"external_cited_by_count":null,"distinct_field_count":4,"first_pith_cited_at":"2026-04-12T14:57:52+00:00","last_pith_cited_at":"2026-07-06T17:59:58+00:00","author_build_status":"not_needed","summary_status":"needed","contexts_status":"needed","graph_status":"needed","ask_index_status":"not_needed","reader_status":"not_needed","recognition_status":"not_needed","updated_at":"2026-08-21T20:40:01.542273+00:00","tier_text":"hub"},"tier":"hub","role_counts":[{"context_role":"background","n":1}],"polarity_counts":[{"context_polarity":"background","n":1}],"runs":{},"summary":{},"graph":{},"authors":[]}}