{"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/unleashing-the-power-of-pre-trained-language","title":"Unleashing the Power of Pre-trained Language Models for Offline Reinforcement Learning","arxiv_id":"2310.20587","date":"2023-10-31","proceeding":null,"authors":["Ruizhe Shi","Yuyao Liu","Yanjie Ze","Simon S. Du","Huazhe Xu"],"abstract":"Offline reinforcement learning (RL) aims to find a near-optimal policy using pre-collected datasets. In real-world scenarios, data collection could be costly and risky; therefore, offline RL becomes particularly challenging when the in-domain data is limited. Given recent advances in Large Language Models (LLMs) and their few-shot learning prowess, this paper introduces $\\textbf{La}$nguage Models for $\\textbf{Mo}$tion Control ($\\textbf{LaMo}$), a general framework based on Decision Transformers to effectively use pre-trained Language Models (LMs) for offline RL. Our framework highlights four crucial components: (1) Initializing Decision Transformers with sequentially pre-trained LMs, (2) employing the LoRA fine-tuning method, in contrast to full-weight fine-tuning, to combine the pre-trained knowledge from LMs and in-domain knowledge effectively, (3) using the non-linear MLP transformation instead of linear projections, to generate embeddings, and (4) integrating an auxiliary language prediction loss during fine-tuning to stabilize the LMs and retain their original abilities on languages. Empirical results indicate $\\textbf{LaMo}$ achieves excellent performance in sparse-reward tasks and closes the gap between value-based offline RL methods and decision transformers in dense-reward tasks. In particular, our method demonstrates superior performance in scenarios with limited data samples.","url_abs":"https://arxiv.org/abs/2310.20587v5","url_pdf":"https://arxiv.org/pdf/2310.20587v5.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":"unleashing-the-power-of-pre-trained-language","repo_url":"https://github.com/srzer/LaMo-2023","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"few-shot-learning","task_name":"Few-Shot Learning"},{"task_slug":"offline-rl","task_name":"Offline RL"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2310.20587","atlas_url":"https://app.syntology.ai/?focus=2310.20587","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2310.20587"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+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/srzer/LaMo-2023","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran_fixture":2,"ran_honours":1,"ran":4,"unverified":2},"by_repo_kind":{"official":{"samples":9,"ran":7,"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":"0161b27cbe3cc58d","entry":"discount_cumsum","repo":"srzer/LaMo-2023","repo_kind":"official","path":"experiment-d4rl/experiment.py","file_url":"https://github.com/srzer/LaMo-2023/blob/HEAD/experiment-d4rl/experiment.py","link_basis":"harvester_set","language":"python","status":"ran_fixture","verification_level":2,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"0161b27cbe3cc58d"}},{"code_sha256_prefix":"a7275f13d13af4f4","entry":"cnn","repo":"srzer/LaMo-2023","repo_kind":"official","path":"experiment-atari/model.py","file_url":"https://github.com/srzer/LaMo-2023/blob/HEAD/experiment-atari/model.py","link_basis":"plan_row","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"a7275f13d13af4f4"}},{"code_sha256_prefix":"46b2ccb71f183c99","entry":"get_cnn_output_dim","repo":"srzer/LaMo-2023","repo_kind":"official","path":"experiment-atari/model.py","file_url":"https://github.com/srzer/LaMo-2023/blob/HEAD/experiment-atari/model.py","link_basis":"plan_row","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"46b2ccb71f183c99"}},{"code_sha256_prefix":"c2615640a56197af","entry":"get_log_dict","repo":"srzer/LaMo-2023","repo_kind":"official","path":"experiment-atari/utils.py","file_url":"https://github.com/srzer/LaMo-2023/blob/HEAD/experiment-atari/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"c2615640a56197af"}},{"code_sha256_prefix":"14c863708fff6f5b","entry":"get_normalized_score","repo":"srzer/LaMo-2023","repo_kind":"official","path":"experiment-atari/get_hns.py","file_url":"https://github.com/srzer/LaMo-2023/blob/HEAD/experiment-atari/get_hns.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"14c863708fff6f5b"}},{"code_sha256_prefix":"d2a121da9fc696b6","entry":"get_optimizer","repo":"srzer/LaMo-2023","repo_kind":"official","path":"experiment-d4rl/utils.py","file_url":"https://github.com/srzer/LaMo-2023/blob/HEAD/experiment-d4rl/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"d2a121da9fc696b6"}},{"code_sha256_prefix":"c89110bdadd7372d","entry":"get_space_shape","repo":"srzer/LaMo-2023","repo_kind":"official","path":"experiment-atari/utils.py","file_url":"https://github.com/srzer/LaMo-2023/blob/HEAD/experiment-atari/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"c89110bdadd7372d"}},{"code_sha256_prefix":"08d865dc3288a3fc","entry":"get_custom_gpt2_model","repo":"srzer/LaMo-2023","repo_kind":"official","path":"experiment-atari/get_custom_gpt2.py","file_url":"https://github.com/srzer/LaMo-2023/blob/HEAD/experiment-atari/get_custom_gpt2.py","link_basis":"first_harvest_node","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":"08d865dc3288a3fc"}},{"code_sha256_prefix":"a42366e534210e01","entry":"get_dataset","repo":"srzer/LaMo-2023","repo_kind":"official","path":"experiment-atari/get_nlp_datasets.py","file_url":"https://github.com/srzer/LaMo-2023/blob/HEAD/experiment-atari/get_nlp_datasets.py","link_basis":"first_harvest_node","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":"a42366e534210e01"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}