{"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/toolsandbox-a-stateful-conversational","title":"ToolSandbox: A Stateful, Conversational, Interactive Evaluation Benchmark for LLM Tool Use Capabilities","arxiv_id":"2408.04682","date":"2024-08-08","proceeding":null,"authors":["Jiarui Lu","Thomas Holleis","Yizhe Zhang","Bernhard Aumayer","Feng Nan","Felix Bai","Shuang Ma","Shen Ma","Mengyu Li","Guoli Yin","ZiRui Wang","Ruoming Pang"],"abstract":"Recent large language models (LLMs) advancements sparked a growing research interest in tool assisted LLMs solving real-world challenges, which calls for comprehensive evaluation of tool-use capabilities. While previous works focused on either evaluating over stateless web services (RESTful API), based on a single turn user prompt, or an off-policy dialog trajectory, ToolSandbox includes stateful tool execution, implicit state dependencies between tools, a built-in user simulator supporting on-policy conversational evaluation and a dynamic evaluation strategy for intermediate and final milestones over an arbitrary trajectory. We show that open source and proprietary models have a significant performance gap, and complex tasks like State Dependency, Canonicalization and Insufficient Information defined in ToolSandbox are challenging even the most capable SOTA LLMs, providing brand-new insights into tool-use LLM capabilities. ToolSandbox evaluation framework is released at https://github.com/apple/ToolSandbox","url_abs":"https://arxiv.org/abs/2408.04682v2","url_pdf":"https://arxiv.org/pdf/2408.04682v2.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":"toolsandbox-a-stateful-conversational","repo_url":"https://github.com/apple/toolsandbox","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2408.04682","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2408.04682"}},"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/apple/toolsandbox","reach":{"status":"ok","spdx":"NOASSERTION"}}],"summary":{"ran":17},"by_repo_kind":{"official":{"samples":17,"ran":17,"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":17,"samples":[{"code_sha256_prefix":"e5749d5e4cb835b4","entry":"column_exact_match_similarity","repo":"apple/toolsandbox","repo_kind":"official","path":"tool_sandbox/common/evaluation.py","file_url":"https://github.com/apple/toolsandbox/blob/HEAD/tool_sandbox/common/evaluation.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"e5749d5e4cb835b4"}},{"code_sha256_prefix":"9c0df2044124dda9","entry":"column_one_similarity","repo":"apple/toolsandbox","repo_kind":"official","path":"tool_sandbox/common/evaluation.py","file_url":"https://github.com/apple/toolsandbox/blob/HEAD/tool_sandbox/common/evaluation.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"9c0df2044124dda9"}},{"code_sha256_prefix":"f70d75ab431ddedb","entry":"default_value_normalization","repo":"apple/toolsandbox","repo_kind":"official","path":"tool_sandbox/common/tool_trace_extractors.py","file_url":"https://github.com/apple/toolsandbox/blob/HEAD/tool_sandbox/common/tool_trace_extractors.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"f70d75ab431ddedb"}},{"code_sha256_prefix":"ad55567b6b0b9a09","entry":"exact_match_filter_dataframe","repo":"apple/toolsandbox","repo_kind":"official","path":"tool_sandbox/common/utils.py","file_url":"https://github.com/apple/toolsandbox/blob/HEAD/tool_sandbox/common/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"ad55567b6b0b9a09"}},{"code_sha256_prefix":"2b077b8d2ea00b28","entry":"extract_aggregated_stats","repo":"apple/toolsandbox","repo_kind":"official","path":"tool_sandbox/analysis/analysis.py","file_url":"https://github.com/apple/toolsandbox/blob/HEAD/tool_sandbox/analysis/analysis.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"2b077b8d2ea00b28"}},{"code_sha256_prefix":"019d435db27eed33","entry":"extract_meta_stats","repo":"apple/toolsandbox","repo_kind":"official","path":"tool_sandbox/analysis/analysis.py","file_url":"https://github.com/apple/toolsandbox/blob/HEAD/tool_sandbox/analysis/analysis.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"019d435db27eed33"}},{"code_sha256_prefix":"547c8e5a97a0b10d","entry":"extract_scenario_results","repo":"apple/toolsandbox","repo_kind":"official","path":"tool_sandbox/analysis/data_loading.py","file_url":"https://github.com/apple/toolsandbox/blob/HEAD/tool_sandbox/analysis/data_loading.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"547c8e5a97a0b10d"}},{"code_sha256_prefix":"25e1f8ca5fb3de8e","entry":"get_scenario_artifacts_path","repo":"apple/toolsandbox","repo_kind":"official","path":"tool_sandbox/analysis/data_loading.py","file_url":"https://github.com/apple/toolsandbox/blob/HEAD/tool_sandbox/analysis/data_loading.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"25e1f8ca5fb3de8e"}},{"code_sha256_prefix":"098e95f00f0b3e5a","entry":"get_scrambled_tool_names","repo":"apple/toolsandbox","repo_kind":"official","path":"tool_sandbox/common/tool_discovery.py","file_url":"https://github.com/apple/toolsandbox/blob/HEAD/tool_sandbox/common/tool_discovery.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"098e95f00f0b3e5a"}},{"code_sha256_prefix":"1ba53634bb0b8cce","entry":"load_result_summary","repo":"apple/toolsandbox","repo_kind":"official","path":"tool_sandbox/analysis/data_loading.py","file_url":"https://github.com/apple/toolsandbox/blob/HEAD/tool_sandbox/analysis/data_loading.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"1ba53634bb0b8cce"}},{"code_sha256_prefix":"f4aa2b10325a6e7f","entry":"openai_tool_call_to_python_code","repo":"apple/toolsandbox","repo_kind":"official","path":"tool_sandbox/common/message_conversion.py","file_url":"https://github.com/apple/toolsandbox/blob/HEAD/tool_sandbox/common/message_conversion.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"f4aa2b10325a6e7f"}},{"code_sha256_prefix":"247415c1aa2c5c82","entry":"python_code_to_openai_tool_call","repo":"apple/toolsandbox","repo_kind":"official","path":"tool_sandbox/common/message_conversion.py","file_url":"https://github.com/apple/toolsandbox/blob/HEAD/tool_sandbox/common/message_conversion.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"247415c1aa2c5c82"}},{"code_sha256_prefix":"55891a36035c3083","entry":"range_filter_dataframe","repo":"apple/toolsandbox","repo_kind":"official","path":"tool_sandbox/common/utils.py","file_url":"https://github.com/apple/toolsandbox/blob/HEAD/tool_sandbox/common/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"55891a36035c3083"}},{"code_sha256_prefix":"32b937d34301c112","entry":"result_to_timestamp0_extractor","repo":"apple/toolsandbox","repo_kind":"official","path":"tool_sandbox/common/tool_trace_extractors.py","file_url":"https://github.com/apple/toolsandbox/blob/HEAD/tool_sandbox/common/tool_trace_extractors.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"32b937d34301c112"}},{"code_sha256_prefix":"f27d06f591963291","entry":"search_weather_around_lat_lon_temperature_extractor","repo":"apple/toolsandbox","repo_kind":"official","path":"tool_sandbox/common/tool_trace_extractors.py","file_url":"https://github.com/apple/toolsandbox/blob/HEAD/tool_sandbox/common/tool_trace_extractors.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"f27d06f591963291"}},{"code_sha256_prefix":"8853c809e4877946","entry":"subsequence_filter_dataframe","repo":"apple/toolsandbox","repo_kind":"official","path":"tool_sandbox/common/utils.py","file_url":"https://github.com/apple/toolsandbox/blob/HEAD/tool_sandbox/common/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"8853c809e4877946"}},{"code_sha256_prefix":"a0ef4c511908e14d","entry":"typechecked","repo":"apple/toolsandbox","repo_kind":"official","path":"tool_sandbox/common/validators.py","file_url":"https://github.com/apple/toolsandbox/blob/HEAD/tool_sandbox/common/validators.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"a0ef4c511908e14d"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}