{"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/arxiv-2602-01365","title":"When Domains Interact: Asymmetric and Order-Sensitive Cross-Domain Effects in Reinforcement Learning for Reasoning","arxiv_id":"2602.01365","date":"2026-02-01","proceeding":null,"authors":["Wang Yang","Shouren Wang","Chaoda Song","Chuang Ma","Xinpeng Li","Nengbo Wang","Kaixiong Zhou","Vipin Chaudhary","Xiaotian Han"],"abstract":"Group Relative Policy Optimization (GRPO) has become a key technique for improving reasoning abilities in large language models, yet its behavior under different domain sequencing strategies is poorly understood. In particular, the impact of sequential (one domain at a time) versus mixed-domain (multiple domain at a time) training in GRPO has not been systematically studied. We provide the first systematic analysis of training-order effects across math, science, logic, and puzzle reasoning tasks. We found (1) single-domain generalization is highly asymmetric: training on other domains improves math reasoning by approximately 25\\% accuracy, while yielding negligible transfer to logic and puzzle; (2) cross-domain interactions are highly order-dependent: training in the order math$\\rightarrow$science achieves 83\\% / 41\\% accuracy on math / science, while reversing the order to science$\\rightarrow$math degrades performance to 77\\% / 25\\%; (3) no single strategy is universally optimal in multi-domain training: sequential training favors math (up to 84\\%), mixed training favors science and logic, and poor ordering can incur large performance gaps (from 70\\% to 56\\%). Overall, our findings demonstrate that GRPO under multi-domain settings exhibits pronounced asymmetry, order sensitivity, and strategy dependence, highlighting the necessity of domain-aware and order-aware training design.","url_abs":"https://arxiv.org/abs/2602.01365","url_pdf":"https://arxiv.org/pdf/2602.01365","source":{"archive":null,"snapshot":"2025-07-28","note":"not in the Papers with Code archive (frozen at the snapshot)","row_kind":"graph","title_abstract_authors_date":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license)"},"code_links":[],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2602.01365","atlas_url":"https://app.syntology.ai/?focus=2602.01365","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2602.01365"}},"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":[{"mentioned_in_github":null,"is_official":null,"provenance":"deterministic:regex_extraction","mentioned_in_paper":null,"url":"https://github.com/uservan/cross_domain","reach":null}],"summary":{"ran":2,"ran_draft_wrong":1,"unverified":9},"by_repo_kind":{"found_in_text":{"samples":12,"ran":3,"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":12,"samples":[{"code_sha256_prefix":"f220595adfc5e83c","entry":"compute_ce_dpo_loss_rm","repo":"uservan/cross_domain","repo_kind":"found_in_text","path":"verl/recipe/prime/prime_core_algos.py","file_url":"https://github.com/uservan/cross_domain/blob/HEAD/verl/recipe/prime/prime_core_algos.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"f220595adfc5e83c"}},{"code_sha256_prefix":"4d91e4383716e42c","entry":"compute_detach_dpo_loss_rm","repo":"uservan/cross_domain","repo_kind":"found_in_text","path":"verl/recipe/prime/prime_core_algos.py","file_url":"https://github.com/uservan/cross_domain/blob/HEAD/verl/recipe/prime/prime_core_algos.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"4d91e4383716e42c"}},{"code_sha256_prefix":"6ead68c2fb69f843","entry":"convert_function_to_class_method","repo":"uservan/cross_domain","repo_kind":"found_in_text","path":"verify/score/livecodebench/compute_score.py","file_url":"https://github.com/uservan/cross_domain/blob/HEAD/verify/score/livecodebench/compute_score.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"6ead68c2fb69f843"}},{"code_sha256_prefix":"8894398260de443f","entry":"check_correctness","repo":"uservan/cross_domain","repo_kind":"found_in_text","path":"verify/score/science.py","file_url":"https://github.com/uservan/cross_domain/blob/HEAD/verify/score/science.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"8894398260de443f"}},{"code_sha256_prefix":"5aefbfebf64faf9c","entry":"compute_score","repo":"uservan/cross_domain","repo_kind":"found_in_text","path":"verify/score/gsm8k.py","file_url":"https://github.com/uservan/cross_domain/blob/HEAD/verify/score/gsm8k.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"5aefbfebf64faf9c"}},{"code_sha256_prefix":"39c5fae4c67fd407","entry":"compute_score","repo":"uservan/cross_domain","repo_kind":"found_in_text","path":"verify/score/puzzle.py","file_url":"https://github.com/uservan/cross_domain/blob/HEAD/verify/score/puzzle.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"39c5fae4c67fd407"}},{"code_sha256_prefix":"d16fd6d7133bfae7","entry":"extract_single_letter","repo":"uservan/cross_domain","repo_kind":"found_in_text","path":"verify/score/science.py","file_url":"https://github.com/uservan/cross_domain/blob/HEAD/verify/score/science.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"d16fd6d7133bfae7"}},{"code_sha256_prefix":"871a5c13b3f8cedb","entry":"extract_solution","repo":"uservan/cross_domain","repo_kind":"found_in_text","path":"verify/score/gsm8k.py","file_url":"https://github.com/uservan/cross_domain/blob/HEAD/verify/score/gsm8k.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"871a5c13b3f8cedb"}},{"code_sha256_prefix":"c39f94fe0e1c8d31","entry":"extract_solution","repo":"uservan/cross_domain","repo_kind":"found_in_text","path":"verify/score/kk.py","file_url":"https://github.com/uservan/cross_domain/blob/HEAD/verify/score/kk.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"c39f94fe0e1c8d31"}},{"code_sha256_prefix":"4bc15242379ac2db","entry":"get_multiple_choice_answer","repo":"uservan/cross_domain","repo_kind":"found_in_text","path":"verify/score/science.py","file_url":"https://github.com/uservan/cross_domain/blob/HEAD/verify/score/science.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"4bc15242379ac2db"}},{"code_sha256_prefix":"c7c851200e535b49","entry":"parse_model_answer","repo":"uservan/cross_domain","repo_kind":"found_in_text","path":"verify/score/kk.py","file_url":"https://github.com/uservan/cross_domain/blob/HEAD/verify/score/kk.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"c7c851200e535b49"}},{"code_sha256_prefix":"152e200c09bfeb4a","entry":"parse_solution_text_format","repo":"uservan/cross_domain","repo_kind":"found_in_text","path":"verify/score/kk.py","file_url":"https://github.com/uservan/cross_domain/blob/HEAD/verify/score/kk.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"152e200c09bfeb4a"}}]},"arxiv_metadata":{"licence":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license)","fields":["title","abstract","authors","date"],"primary_category":"cs.LG","source":"arxiv_2026.jsonl"},"syntology_extracted_results":null}