{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:2O2ZSRUNAFMH7UGPORUNN3XS5W","short_pith_number":"pith:2O2ZSRUN","schema_version":"1.0","canonical_sha256":"d3b599468d01587fd0cf7468d6eef2ed90288fde8cb9466bbb208fc99dad6d95","source":{"kind":"arxiv","id":"2505.15469","version":1},"attestation_state":"computed","paper":{"title":"A Qualitative Investigation into LLM-Generated Multilingual Code Comments and Automatic Evaluation Metrics","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.SE","authors_text":"Arie van Deursen, Gopal-Raj Panchu, Jonathan Katzy, Maksym Ziemlewski, Maliheh Izadi, Paris Loizides, Sander Vermeulen, Yongcheng Huang","submitted_at":"2025-05-21T12:45:49Z","abstract_excerpt":"Large Language Models are essential coding assistants, yet their training is predominantly English-centric. In this study, we evaluate the performance of code language models in non-English contexts, identifying challenges in their adoption and integration into multilingual workflows. We conduct an open-coding study to analyze errors in code comments generated by five state-of-the-art code models, CodeGemma, CodeLlama, CodeQwen1.5, GraniteCode, and StarCoder2 across five natural languages: Chinese, Dutch, English, Greek, and Polish. Our study yields a dataset of 12,500 labeled generations, whi"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2505.15469","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.SE","submitted_at":"2025-05-21T12:45:49Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"13aad9c40952e7162b4c2b53634857342fbfd3a0e42618f3e1fe737f2e752d41","abstract_canon_sha256":"37cc7e42e387e90b6ecb558be8464630960bc0e41972dedcc94c8e36a22baf47"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:06:46.285888Z","signature_b64":"Wa5gHQ0a+KQqxMDOmvDlUPiuLTvRl8Z2SUSHAVMp5fuXJJ1DNuA+LCpkmavCKzIuK/IQvCPN60TYbcPgDixWAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d3b599468d01587fd0cf7468d6eef2ed90288fde8cb9466bbb208fc99dad6d95","last_reissued_at":"2026-07-05T11:06:46.285404Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:06:46.285404Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Qualitative Investigation into LLM-Generated Multilingual Code Comments and Automatic Evaluation Metrics","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.SE","authors_text":"Arie van Deursen, Gopal-Raj Panchu, Jonathan Katzy, Maksym Ziemlewski, Maliheh Izadi, Paris Loizides, Sander Vermeulen, Yongcheng Huang","submitted_at":"2025-05-21T12:45:49Z","abstract_excerpt":"Large Language Models are essential coding assistants, yet their training is predominantly English-centric. In this study, we evaluate the performance of code language models in non-English contexts, identifying challenges in their adoption and integration into multilingual workflows. We conduct an open-coding study to analyze errors in code comments generated by five state-of-the-art code models, CodeGemma, CodeLlama, CodeQwen1.5, GraniteCode, and StarCoder2 across five natural languages: Chinese, Dutch, English, Greek, and Polish. Our study yields a dataset of 12,500 labeled generations, whi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.15469","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2505.15469/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2505.15469","created_at":"2026-07-05T11:06:46.285467+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.15469v1","created_at":"2026-07-05T11:06:46.285467+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.15469","created_at":"2026-07-05T11:06:46.285467+00:00"},{"alias_kind":"pith_short_12","alias_value":"2O2ZSRUNAFMH","created_at":"2026-07-05T11:06:46.285467+00:00"},{"alias_kind":"pith_short_16","alias_value":"2O2ZSRUNAFMH7UGP","created_at":"2026-07-05T11:06:46.285467+00:00"},{"alias_kind":"pith_short_8","alias_value":"2O2ZSRUN","created_at":"2026-07-05T11:06:46.285467+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2506.08980","citing_title":"AdaDec: A Uncertainty-Guided Lookahead Decoding Framework for LLM-Based Code Generation","ref_index":35,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2O2ZSRUNAFMH7UGPORUNN3XS5W","json":"https://pith.science/pith/2O2ZSRUNAFMH7UGPORUNN3XS5W.json","graph_json":"https://pith.science/api/pith-number/2O2ZSRUNAFMH7UGPORUNN3XS5W/graph.json","events_json":"https://pith.science/api/pith-number/2O2ZSRUNAFMH7UGPORUNN3XS5W/events.json","paper":"https://pith.science/paper/2O2ZSRUN"},"agent_actions":{"view_html":"https://pith.science/pith/2O2ZSRUNAFMH7UGPORUNN3XS5W","download_json":"https://pith.science/pith/2O2ZSRUNAFMH7UGPORUNN3XS5W.json","view_paper":"https://pith.science/paper/2O2ZSRUN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.15469&json=true","fetch_graph":"https://pith.science/api/pith-number/2O2ZSRUNAFMH7UGPORUNN3XS5W/graph.json","fetch_events":"https://pith.science/api/pith-number/2O2ZSRUNAFMH7UGPORUNN3XS5W/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2O2ZSRUNAFMH7UGPORUNN3XS5W/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2O2ZSRUNAFMH7UGPORUNN3XS5W/action/storage_attestation","attest_author":"https://pith.science/pith/2O2ZSRUNAFMH7UGPORUNN3XS5W/action/author_attestation","sign_citation":"https://pith.science/pith/2O2ZSRUNAFMH7UGPORUNN3XS5W/action/citation_signature","submit_replication":"https://pith.science/pith/2O2ZSRUNAFMH7UGPORUNN3XS5W/action/replication_record"}},"created_at":"2026-07-05T11:06:46.285467+00:00","updated_at":"2026-07-05T11:06:46.285467+00:00"}