{"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/agent-models-internalizing-chain-of-action","title":"Agent models: Internalizing Chain-of-Action Generation into Reasoning models","arxiv_id":"2503.06580","date":"2025-03-09","proceeding":null,"authors":["Yuxiang Zhang","YuQi Yang","Jiangming Shu","Xinyan Wen","Jitao Sang"],"abstract":"Traditional agentic workflows rely on external prompts to manage interactions with tools and the environment, which limits the autonomy of reasoning models. We position \\emph{Large Agent Models (LAMs)} that internalize the generation of \\emph{Chain-of-Action (CoA)}, enabling the model to autonomously decide when and how to use external tools. Our proposed AutoCoA framework combines supervised fine-tuning (SFT) and reinforcement learning (RL), allowing the model to seamlessly switch between reasoning and action while efficiently managing environment interactions. Main components include step-level action triggering, trajectory-level CoA optimization, and an internal world model to reduce real-environment interaction costs. Evaluations on open-domain QA tasks demonstrate that AutoCoA-trained agent models significantly outperform ReAct-based workflows in task completion, especially in tasks that require long-term reasoning and multi-step actions. Code and dataset are available at https://github.com/ADaM-BJTU/AutoCoA","url_abs":"https://arxiv.org/abs/2503.06580v1","url_pdf":"https://arxiv.org/pdf/2503.06580v1.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":"agent-models-internalizing-chain-of-action","repo_url":"https://github.com/adam-bjtu/autocoa","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"action-generation","task_name":"Action Generation"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2503.06580","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2503.06580"}},"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/adam-bjtu/autocoa","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran_fixture":1,"ran_draft_wrong":1,"unverified":12},"by_repo_kind":{"official":{"samples":14,"ran":2,"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":"21331a58f93375e4","entry":"union_tensor_dict","repo":"adam-bjtu/autocoa","repo_kind":"official","path":"RL_training/verl/protocol.py","file_url":"https://github.com/adam-bjtu/autocoa/blob/HEAD/RL_training/verl/protocol.py","link_basis":"harvester_set","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"DEP_MISSING","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"21331a58f93375e4"}},{"code_sha256_prefix":"25f0ea3f460f6ce1","entry":"unpad_dataproto","repo":"adam-bjtu/autocoa","repo_kind":"official","path":"RL_training/verl/protocol.py","file_url":"https://github.com/adam-bjtu/autocoa/blob/HEAD/RL_training/verl/protocol.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"25f0ea3f460f6ce1"}},{"code_sha256_prefix":"f4a300d49c56f7e4","entry":"check_search_token","repo":"adam-bjtu/autocoa","repo_kind":"official","path":"evaluation/eval_retrieval.py","file_url":"https://github.com/adam-bjtu/autocoa/blob/HEAD/evaluation/eval_retrieval.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":"f4a300d49c56f7e4"}},{"code_sha256_prefix":"aa0d5e49726f39b2","entry":"count_search_nums","repo":"adam-bjtu/autocoa","repo_kind":"official","path":"evaluation/eval_retrieval.py","file_url":"https://github.com/adam-bjtu/autocoa/blob/HEAD/evaluation/eval_retrieval.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":"aa0d5e49726f39b2"}},{"code_sha256_prefix":"6d99fd0ee8832cb2","entry":"extract_boxed_answer","repo":"adam-bjtu/autocoa","repo_kind":"official","path":"evaluation/eval_retrieval.py","file_url":"https://github.com/adam-bjtu/autocoa/blob/HEAD/evaluation/eval_retrieval.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":"6d99fd0ee8832cb2"}},{"code_sha256_prefix":"fe8edf17e376de99","entry":"format_results","repo":"adam-bjtu/autocoa","repo_kind":"official","path":"models/wiki_engine.py","file_url":"https://github.com/adam-bjtu/autocoa/blob/HEAD/models/wiki_engine.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":"fe8edf17e376de99"}},{"code_sha256_prefix":"ddc4de55fc7f4e3b","entry":"generate_chat_completion","repo":"adam-bjtu/autocoa","repo_kind":"official","path":"models/vllm_server.py","file_url":"https://github.com/adam-bjtu/autocoa/blob/HEAD/models/vllm_server.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":"ddc4de55fc7f4e3b"}},{"code_sha256_prefix":"c89ca5a383e7a0bc","entry":"generate_completion","repo":"adam-bjtu/autocoa","repo_kind":"official","path":"models/vllm_server.py","file_url":"https://github.com/adam-bjtu/autocoa/blob/HEAD/models/vllm_server.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":"c89ca5a383e7a0bc"}},{"code_sha256_prefix":"660713a8fa047736","entry":"get_generation_config","repo":"adam-bjtu/autocoa","repo_kind":"official","path":"RL_training/verl/utils/model.py","file_url":"https://github.com/adam-bjtu/autocoa/blob/HEAD/RL_training/verl/utils/model.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":"660713a8fa047736"}},{"code_sha256_prefix":"fe79c459a3e4cc64","entry":"get_huggingface_actor_config","repo":"adam-bjtu/autocoa","repo_kind":"official","path":"RL_training/verl/utils/model.py","file_url":"https://github.com/adam-bjtu/autocoa/blob/HEAD/RL_training/verl/utils/model.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":"fe79c459a3e4cc64"}},{"code_sha256_prefix":"59bb946c117d4e32","entry":"get_search_results","repo":"adam-bjtu/autocoa","repo_kind":"official","path":"models/wiki_engine.py","file_url":"https://github.com/adam-bjtu/autocoa/blob/HEAD/models/wiki_engine.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":"59bb946c117d4e32"}},{"code_sha256_prefix":"76f3c749a0e655f2","entry":"get_weight_loader","repo":"adam-bjtu/autocoa","repo_kind":"official","path":"RL_training/verl/models/weight_loader_registry.py","file_url":"https://github.com/adam-bjtu/autocoa/blob/HEAD/RL_training/verl/models/weight_loader_registry.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":"76f3c749a0e655f2"}},{"code_sha256_prefix":"98a43410c5681732","entry":"normalize_answer","repo":"adam-bjtu/autocoa","repo_kind":"official","path":"evaluation/metrics.py","file_url":"https://github.com/adam-bjtu/autocoa/blob/HEAD/evaluation/metrics.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":"98a43410c5681732"}},{"code_sha256_prefix":"3b15e2ac7497c441","entry":"squeeze","repo":"adam-bjtu/autocoa","repo_kind":"official","path":"RL_training/verl/utils/model.py","file_url":"https://github.com/adam-bjtu/autocoa/blob/HEAD/RL_training/verl/utils/model.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":"3b15e2ac7497c441"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}