{"modele":[{"model":"models/gemini-3.1-flash-lite","workflow":"analyst"}],"wersja":"aipoc-lab/pochodzenie-0.1","zrodla":[{"tagi":["kat/tech","pod/anthropic","pod/openai","tem/agents","tem/automation","tem/llm","typ/analysis"],"tytul":"Uczenie przez wzmocnienie (RL) w modelach językowych","zebrano":"2026-09-20T02:50:45.898081+00:00","sygnal_id":14857}],"lancuch":[{"na":"new","aktor":"analyst","kiedy":"2026-09-20T04:45:08.30969+00:00","powod":"study z notatki uczenie-ze-wzmocnieniem wersja 6 (500 sygnalow zrodlowych)"},{"z":"new","na":"approved","aktor":"pm","kiedy":"2026-09-20T09:52:56.83512+00:00","powod":"apr"},{"z":"approved","na":"published","aktor":"publisher","kiedy":"2026-09-20T10:00:00.460754+00:00","powod":"payload doc 316"}],"artefakt":{"id":14862,"tagi":["kat/tech","pod/anthropic","pod/openai","tem/agents","tem/automation","tem/llm","typ/analysis"],"tytul":"Uczenie przez wzmocnienie w nowoczesnych systemach agentowych","rodzaj":"study","status":"published","utworzony":"2026-09-20T04:45:08.30969+00:00","dokument_cms":"316"},"przebiegi":[{"start":"2026-09-20T02:50:39.384882+00:00","koniec":"2026-09-20T02:50:46.259203+00:00","run_id":1149,"status":"ok","workflow":"note"},{"start":"2026-09-20T04:45:00.177298+00:00","koniec":"2026-09-20T04:45:08.369762+00:00","run_id":1169,"status":"ok","workflow":"analyst"}]}