{"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/remedy-learning-machine-translation","title":"Remedy: Learning Machine Translation Evaluation from Human Preferences with Reward Modeling","arxiv_id":"2504.13630","date":"2025-04-18","proceeding":null,"authors":["Shaomu Tan","Christof Monz"],"abstract":"A key challenge in MT evaluation is the inherent noise and inconsistency of human ratings. Regression-based neural metrics struggle with this noise, while prompting LLMs shows promise at system-level evaluation but performs poorly at segment level. In this work, we propose ReMedy, a novel MT metric framework that reformulates translation evaluation as a reward modeling task. Instead of regressing on imperfect human ratings directly, ReMedy learns relative translation quality using pairwise preference data, resulting in a more reliable evaluation. In extensive experiments across WMT22-24 shared tasks (39 language pairs, 111 MT systems), ReMedy achieves state-of-the-art performance at both segment- and system-level evaluation. Specifically, ReMedy-9B surpasses larger WMT winners and massive closed LLMs such as MetricX-13B, XCOMET-Ensemble, GEMBA-GPT-4, PaLM-540B, and finetuned PaLM2. Further analyses demonstrate that ReMedy delivers superior capability in detecting translation errors and evaluating low-quality translations.","url_abs":"https://arxiv.org/abs/2504.13630v1","url_pdf":"https://arxiv.org/pdf/2504.13630v1.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":"remedy-learning-machine-translation","repo_url":"https://github.com/Smu-Tan/Remedy","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"machine-translation","task_name":"Machine Translation"},{"task_slug":"translation","task_name":"Translation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2504.13630","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2504.13630"}},"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":"deterministic:regex_extraction","url":"https://github.com/Smu-Tan/Remedy","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"unverified":13},"by_repo_kind":{"official":{"samples":13,"ran":0,"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":"2f9cee62218224c4","entry":"collect_all_scores","repo":"Smu-Tan/Remedy","repo_kind":"official","path":"remedy/toolbox/calibration.py","file_url":"https://github.com/Smu-Tan/Remedy/blob/HEAD/remedy/toolbox/calibration.py","link_basis":"harvester_set","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":"2f9cee62218224c4"}},{"code_sha256_prefix":"d8cb40f7ba05a108","entry":"create_dataframe","repo":"Smu-Tan/Remedy","repo_kind":"official","path":"remedy/toolbox/eval_wmt/extract_result_wmt22.py","file_url":"https://github.com/Smu-Tan/Remedy/blob/HEAD/remedy/toolbox/eval_wmt/extract_result_wmt22.py","link_basis":"harvester_set","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":"d8cb40f7ba05a108"}},{"code_sha256_prefix":"09f21f989aaacf1e","entry":"create_dataframe","repo":"Smu-Tan/Remedy","repo_kind":"official","path":"remedy/toolbox/eval_wmt/extract_result_wmt23.py","file_url":"https://github.com/Smu-Tan/Remedy/blob/HEAD/remedy/toolbox/eval_wmt/extract_result_wmt23.py","link_basis":"harvester_set","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":"09f21f989aaacf1e"}},{"code_sha256_prefix":"c2108c8a06f7aff7","entry":"extract_metrics","repo":"Smu-Tan/Remedy","repo_kind":"official","path":"remedy/toolbox/eval_wmt/extract_result_wmt22.py","file_url":"https://github.com/Smu-Tan/Remedy/blob/HEAD/remedy/toolbox/eval_wmt/extract_result_wmt22.py","link_basis":"harvester_set","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":"c2108c8a06f7aff7"}},{"code_sha256_prefix":"78e2a60b74bbece6","entry":"extract_metrics","repo":"Smu-Tan/Remedy","repo_kind":"official","path":"remedy/toolbox/eval_wmt/extract_result_wmt23.py","file_url":"https://github.com/Smu-Tan/Remedy/blob/HEAD/remedy/toolbox/eval_wmt/extract_result_wmt23.py","link_basis":"harvester_set","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":"78e2a60b74bbece6"}},{"code_sha256_prefix":"49ac166cd19925c7","entry":"extract_sys_metrics","repo":"Smu-Tan/Remedy","repo_kind":"official","path":"remedy/toolbox/eval_wmt/extract_result_wmt22.py","file_url":"https://github.com/Smu-Tan/Remedy/blob/HEAD/remedy/toolbox/eval_wmt/extract_result_wmt22.py","link_basis":"harvester_set","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":"49ac166cd19925c7"}},{"code_sha256_prefix":"37ec69329a14a927","entry":"extract_sys_metrics","repo":"Smu-Tan/Remedy","repo_kind":"official","path":"remedy/toolbox/eval_wmt/extract_result_wmt23.py","file_url":"https://github.com/Smu-Tan/Remedy/blob/HEAD/remedy/toolbox/eval_wmt/extract_result_wmt23.py","link_basis":"harvester_set","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":"37ec69329a14a927"}},{"code_sha256_prefix":"f9e4cf195b0484ff","entry":"final_optimized_entropy","repo":"Smu-Tan/Remedy","repo_kind":"official","path":"remedy/toolbox/calibration.py","file_url":"https://github.com/Smu-Tan/Remedy/blob/HEAD/remedy/toolbox/calibration.py","link_basis":"harvester_set","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":"f9e4cf195b0484ff"}},{"code_sha256_prefix":"d1291ad7e7513a2b","entry":"get_full_lang_name","repo":"Smu-Tan/Remedy","repo_kind":"official","path":"remedy/toolbox/languages.py","file_url":"https://github.com/Smu-Tan/Remedy/blob/HEAD/remedy/toolbox/languages.py","link_basis":"harvester_set","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":"d1291ad7e7513a2b"}},{"code_sha256_prefix":"56c142f70a3ca3fe","entry":"is_supported_language","repo":"Smu-Tan/Remedy","repo_kind":"official","path":"remedy/toolbox/languages.py","file_url":"https://github.com/Smu-Tan/Remedy/blob/HEAD/remedy/toolbox/languages.py","link_basis":"harvester_set","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":"56c142f70a3ca3fe"}},{"code_sha256_prefix":"25030bf189a91632","entry":"load_score_from_text","repo":"Smu-Tan/Remedy","repo_kind":"official","path":"remedy/toolbox/eval_wmt/extract_wmt24_esa.py","file_url":"https://github.com/Smu-Tan/Remedy/blob/HEAD/remedy/toolbox/eval_wmt/extract_wmt24_esa.py","link_basis":"harvester_set","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":"25030bf189a91632"}},{"code_sha256_prefix":"ef92151eb4f55698","entry":"load_translation_data","repo":"Smu-Tan/Remedy","repo_kind":"official","path":"remedy/toolbox/score.py","file_url":"https://github.com/Smu-Tan/Remedy/blob/HEAD/remedy/toolbox/score.py","link_basis":"harvester_set","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":"ef92151eb4f55698"}},{"code_sha256_prefix":"9da49703a6c63108","entry":"sigmoid","repo":"Smu-Tan/Remedy","repo_kind":"official","path":"remedy/toolbox/calibration.py","file_url":"https://github.com/Smu-Tan/Remedy/blob/HEAD/remedy/toolbox/calibration.py","link_basis":"harvester_set","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":"9da49703a6c63108"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}