{"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-2606-20629","title":"Specialize Roles, Mix Deployments: Pushing the Cost-Accuracy Frontier of LLM Agent Teams","arxiv_id":"2606.20629","date":"2026-05-28","proceeding":null,"authors":["Yinsicheng Jiang","Liang Cheng","Yeqi Huang","Yufan Zhao","Zhan Lu","Li Dong","Wenda Li","Edoardo Ponti","Luo Mai"],"abstract":"LLM agents are increasingly deployed as multi-role teams, where tasks are divided across specialized roles such as planner, executor, and verifier. In these systems, cost and accuracy are no longer properties of a single model: they depend on which model fills each role and where it is hosted, including API, self-hosted, and hybrid deployment. Existing agentic benchmarks typically evaluate fixed models or fixed agent configurations, and therefore offer limited guidance for cost-accuracy-optimal deployment. We introduce AgentCARD, a role-aware benchmark suite for evaluating LLM agent teams across role assignment and deployment mode. AgentCARD combines a role-decomposed evaluation harness, a unified API/self-hosted cost model, Pareto-frontier analysis, and a Shapley-based diagnostic for identifying role bottlenecks. Our evaluation shows that heterogeneous teams consistently occupy the cost-accuracy frontier. They improve accuracy by up to $44\\%$ over cost-equivalent homogeneous teams, or match the strongest homogeneous team at up to $12\\times$ lower per-task cost through hybrid deployment. We further find that the best role assignment is domain-dependent: some domains are planner-bottlenecked, while others are executor-bottlenecked. Finally, AgentCARD extends beyond planner--executor teams to workflows with additional roles such as verification, and supports continual evaluation as new domains and team structures emerge. Our code is released at: https://github.com/Auto-CAP/AgentCAP","url_abs":"https://arxiv.org/abs/2606.20629","url_pdf":"https://arxiv.org/pdf/2606.20629","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":{"atlas_url":null,"mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2606.20629"}},"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":[{"mentioned_in_github":null,"is_official":null,"provenance":"deterministic:regex_extraction","mentioned_in_paper":null,"url":"https://github.com/Auto-CAP/AgentCAP","reach":null}],"summary":{"ran":6,"unverified":1},"by_repo_kind":{"found_in_text":{"samples":7,"ran":6,"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":"c4a35e2484140104","entry":"apply_llm_judgment","repo":"Auto-CAP/AgentCAP","repo_kind":"found_in_text","path":"agent_cap/run_imo_answerbench_4.py","file_url":"https://github.com/Auto-CAP/AgentCAP/blob/HEAD/agent_cap/run_imo_answerbench_4.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"c4a35e2484140104"}},{"code_sha256_prefix":"f2429b8e80856961","entry":"build_failed_task_result","repo":"Auto-CAP/AgentCAP","repo_kind":"found_in_text","path":"agent_cap/imo_output.py","file_url":"https://github.com/Auto-CAP/AgentCAP/blob/HEAD/agent_cap/imo_output.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"f2429b8e80856961"}},{"code_sha256_prefix":"e58b1915aa9b1fcb","entry":"create_timer","repo":"Auto-CAP/AgentCAP","repo_kind":"found_in_text","path":"agent_cap/core/timer.py","file_url":"https://github.com/Auto-CAP/AgentCAP/blob/HEAD/agent_cap/core/timer.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"e58b1915aa9b1fcb"}},{"code_sha256_prefix":"43e1d98f53cdbc9a","entry":"positive_int","repo":"Auto-CAP/AgentCAP","repo_kind":"found_in_text","path":"agent_cap/imo_output.py","file_url":"https://github.com/Auto-CAP/AgentCAP/blob/HEAD/agent_cap/imo_output.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"43e1d98f53cdbc9a"}},{"code_sha256_prefix":"f4edb838b05216e5","entry":"resolve_model_path","repo":"Auto-CAP/AgentCAP","repo_kind":"found_in_text","path":"agent_cap/run_imo_answerbench_4.py","file_url":"https://github.com/Auto-CAP/AgentCAP/blob/HEAD/agent_cap/run_imo_answerbench_4.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"f4edb838b05216e5"}},{"code_sha256_prefix":"fa2bc096f2863e2b","entry":"tracked_stream","repo":"Auto-CAP/AgentCAP","repo_kind":"found_in_text","path":"agent_cap/imo_output.py","file_url":"https://github.com/Auto-CAP/AgentCAP/blob/HEAD/agent_cap/imo_output.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"fa2bc096f2863e2b"}},{"code_sha256_prefix":"6f5f18826bfcfa69","entry":"evaluate","repo":"Auto-CAP/AgentCAP","repo_kind":"found_in_text","path":"agent_cap/evaluator.py","file_url":"https://github.com/Auto-CAP/AgentCAP/blob/HEAD/agent_cap/evaluator.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"6f5f18826bfcfa69"}}]},"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}