{"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/res-q-evaluating-code-editing-large-language","title":"RES-Q: Evaluating Code-Editing Large Language Model Systems at the Repository Scale","arxiv_id":"2406.16801","date":"2024-06-24","proceeding":null,"authors":["Beck Labash","August Rosedale","Alex Reents","Lucas Negritto","Colin Wiel"],"abstract":"The instruction-following ability of Large Language Models (LLMs) has cultivated a class of LLM-based systems capable of approaching complex tasks such as making edits to large code repositories. Due to the high sensitivity and unpredictability of LLM behavior in response to changes in prompting, robust evaluation tools are needed to drive future iteration of these systems. We propose RES-Q, a natural language instruction-based benchmark for evaluating $\\textbf{R}$epository $\\textbf{E}$diting $\\textbf{S}$ystems, which consists of 100 handcrafted repository editing tasks derived from real GitHub commits. Given an edit instruction and a code repository, RES-Q evaluates an LLM system's ability to interpret the instruction, navigate the repository to gather relevant information, and construct an appropriate edit that satisfies the specified criteria. We argue that evaluating LLMs in this way addresses issues with traditional benchmarks and provides a more holistic assessment of a model's abilities. We evaluate various state-of-the-art LLMs as language agents in a repository-editing system built on Qurrent OS, our language agent development software. Despite their 1% pass@1 performance difference on HumanEval, we find Claude Sonnet 3.5 outperforms GPT-4o by 12% pass@1 on RES-Q, indicating RES-Q's capacity to differentiate model capability as traditional benchmarks approach saturation. We further investigate token efficiency, performance relationships with existing benchmarks, and interesting disparities between closed and open-source LLMs. Code and dataset are available at https://github.com/Qurrent-AI/RES-Q.","url_abs":"https://arxiv.org/abs/2406.16801v2","url_pdf":"https://arxiv.org/pdf/2406.16801v2.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":"res-q-evaluating-code-editing-large-language","repo_url":"https://github.com/qurrent-ai/res-q","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"code-generation","task_name":"Code Generation"},{"task_slug":"humaneval","task_name":"HumanEval"},{"task_slug":"instruction-following","task_name":"Instruction Following"},{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"large-language-model","task_name":"Large Language Model"},{"task_slug":"navigate","task_name":"Navigate"}],"methods":[],"datasets_introduced":[{"slug":"res-q","name":"RES-Q","full_name":"RES-Q: Evaluating Code-Editing Large Language Model Systems at the Repository Scale"}],"methods_introduced":[],"results":[{"leaderboard":"/sota/code-generation-on-res-q","task":"Code Generation","dataset":"RES-Q","model":"QurrentOS-coder + Claude 3.5 Sonnet","rank_in_archive_order":1,"of":9,"metrics":{"pass@1":"58.0"},"uses_additional_data":false},{"leaderboard":"/sota/code-generation-on-res-q","task":"Code Generation","dataset":"RES-Q","model":"QurrentOS-coder + GPT-4o","rank_in_archive_order":2,"of":9,"metrics":{"pass@1":"46.0"},"uses_additional_data":false},{"leaderboard":"/sota/code-generation-on-res-q","task":"Code Generation","dataset":"RES-Q","model":"QurrentOS-coder + GPT-4 Turbo","rank_in_archive_order":3,"of":9,"metrics":{"pass@1":"37.0"},"uses_additional_data":false},{"leaderboard":"/sota/code-generation-on-res-q","task":"Code Generation","dataset":"RES-Q","model":"QurrentOS-coder + Claude 3 Opus","rank_in_archive_order":4,"of":9,"metrics":{"pass@1":"36.0"},"uses_additional_data":false},{"leaderboard":"/sota/code-generation-on-res-q","task":"Code Generation","dataset":"RES-Q","model":"QurrentOS-coder + GPT-4","rank_in_archive_order":5,"of":9,"metrics":{"pass@1":"30.0"},"uses_additional_data":false},{"leaderboard":"/sota/code-generation-on-res-q","task":"Code Generation","dataset":"RES-Q","model":"QurrentOS-coder + Gemini 1.5 Pro","rank_in_archive_order":6,"of":9,"metrics":{"pass@1":"30.0"},"uses_additional_data":false},{"leaderboard":"/sota/code-generation-on-res-q","task":"Code Generation","dataset":"RES-Q","model":"QurrentOS-coder + DeepSeek-Coder-V2","rank_in_archive_order":7,"of":9,"metrics":{"pass@1":"29.0"},"uses_additional_data":false},{"leaderboard":"/sota/code-generation-on-res-q","task":"Code Generation","dataset":"RES-Q","model":"QurrentOS-coder + Llama 3 70b","rank_in_archive_order":8,"of":9,"metrics":{"pass@1":"20.0"},"uses_additional_data":false},{"leaderboard":"/sota/code-generation-on-res-q","task":"Code Generation","dataset":"RES-Q","model":"QurrentOS-coder + Qwen-72B-Instruct","rank_in_archive_order":9,"of":9,"metrics":{"pass@1":"18.0"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2406.16801","atlas_url":"https://app.syntology.ai/?focus=2406.16801","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2406.16801"}},"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/qurrent-ai/res-q","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":6},"by_repo_kind":{"official":{"samples":6,"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":"48c807f6feeb0909","entry":"extract_modified_files","repo":"qurrent-ai/res-q","repo_kind":"official","path":"resq/git/utils.py","file_url":"https://github.com/qurrent-ai/res-q/blob/HEAD/resq/git/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"48c807f6feeb0909"}},{"code_sha256_prefix":"76b18236be6ad0d9","entry":"filter_binary_patch","repo":"qurrent-ai/res-q","repo_kind":"official","path":"resq/git/utils.py","file_url":"https://github.com/qurrent-ai/res-q/blob/HEAD/resq/git/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"76b18236be6ad0d9"}},{"code_sha256_prefix":"e16f1ea45d6f67d6","entry":"format_patch","repo":"qurrent-ai/res-q","repo_kind":"official","path":"resq/git/utils.py","file_url":"https://github.com/qurrent-ai/res-q/blob/HEAD/resq/git/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"e16f1ea45d6f67d6"}},{"code_sha256_prefix":"0751670779a93959","entry":"insert_secret","repo":"qurrent-ai/res-q","repo_kind":"official","path":"resq/testbeds/utils.py","file_url":"https://github.com/qurrent-ai/res-q/blob/HEAD/resq/testbeds/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"0751670779a93959"}},{"code_sha256_prefix":"468a2041e486fed6","entry":"log_run_subprocess","repo":"qurrent-ai/res-q","repo_kind":"official","path":"resq/utils.py","file_url":"https://github.com/qurrent-ai/res-q/blob/HEAD/resq/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":"468a2041e486fed6"}},{"code_sha256_prefix":"1fda66523279af5a","entry":"log_run_subprocess_sync","repo":"qurrent-ai/res-q","repo_kind":"official","path":"resq/utils.py","file_url":"https://github.com/qurrent-ai/res-q/blob/HEAD/resq/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":"1fda66523279af5a"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}