{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/2505-10861","title":"Improving the Data-efficiency of Reinforcement Learning by Warm-starting with LLM","arxiv_id":"2505.10861","date":"2025-05-16","proceeding":null,"authors":["Thang Duong","Minglai Yang","Chicheng Zhang"],"abstract":"We investigate the usage of Large Language Model (LLM) in collecting high-quality data to warm-start Reinforcement Learning (RL) algorithms for learning in some classical Markov Decision Process (MDP) environments. In this work, we focus on using LLM to generate an off-policy dataset that sufficiently covers state-actions visited by optimal policies, then later using an RL algorithm to explore the environment and improve the policy suggested by the LLM. Our algorithm, LORO, can both converge to an optimal policy and have a high sample efficiency thanks to the LLM's good starting policy. On multiple OpenAI Gym environments, such as CartPole and Pendulum, we empirically demonstrate that LORO outperforms baseline algorithms such as pure LLM-based policies, pure RL, and a naive combination of the two, achieving up to $4 \\times$ the cumulative rewards of the pure RL baseline.","url_abs":"https://arxiv.org/abs/2505.10861v1","url_pdf":"https://arxiv.org/pdf/2505.10861v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"2505-10861","repo_url":"https://github.com/duongnhatthang/llamagym","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"large-language-model","task_name":"Large Language Model"},{"task_slug":"openai-gym","task_name":"OpenAI Gym"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"}],"methods":[{"method_slug":"focus","method_name":"Focus"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2505.10861","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2505.10861"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/duongnhatthang/llamagym","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"unverified":5},"by_repo_kind":{"official":{"samples":5,"ran":0,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"5ac85b6d73db33f8","entry":"custom_create_reference_model","repo":"duongnhatthang/llamagym","repo_kind":"official","path":"llamagym/agent.py","file_url":"https://github.com/duongnhatthang/llamagym/blob/HEAD/llamagym/agent.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"5ac85b6d73db33f8"}},{"code_sha256_prefix":"56a1773eaedefa04","entry":"ensure_numpy_array","repo":"duongnhatthang/llamagym","repo_kind":"official","path":"pretrain_from_llm.py","file_url":"https://github.com/duongnhatthang/llamagym/blob/HEAD/pretrain_from_llm.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"56a1773eaedefa04"}},{"code_sha256_prefix":"1a022fa5ea98c3cd","entry":"env_factory","repo":"duongnhatthang/llamagym","repo_kind":"official","path":"env/atari/represented_atari_game.py","file_url":"https://github.com/duongnhatthang/llamagym/blob/HEAD/env/atari/represented_atari_game.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"1a022fa5ea98c3cd"}},{"code_sha256_prefix":"d60a328964c3ddb4","entry":"evaluate_qlearning_with_environment","repo":"duongnhatthang/llamagym","repo_kind":"official","path":"online_main.py","file_url":"https://github.com/duongnhatthang/llamagym/blob/HEAD/online_main.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"d60a328964c3ddb4"}},{"code_sha256_prefix":"64b07d05d404f780","entry":"get_llm_data_paths","repo":"duongnhatthang/llamagym","repo_kind":"official","path":"pretrain_from_llm.py","file_url":"https://github.com/duongnhatthang/llamagym/blob/HEAD/pretrain_from_llm.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"64b07d05d404f780"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}