{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:5O3KPRVQSD3HV2UYXUOVN65LJN","short_pith_number":"pith:5O3KPRVQ","schema_version":"1.0","canonical_sha256":"ebb6a7c6b090f67aea98bd1d56fbab4b470fd73abee2a4da3e60587dc245162b","source":{"kind":"arxiv","id":"2504.08120","version":3},"attestation_state":"computed","paper":{"title":"DeepSeek-R1 vs. o3-mini: How Well can Reasoning LLMs Evaluate MT and Summarization?","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Christian Greisinger, Christoph Leiter, Daniil Larionov, Ran Zhang, Sotaro Takeshita, Steffen Eger, Yanran Chen, Zhipin Wang","submitted_at":"2025-04-10T20:39:18Z","abstract_excerpt":"Reasoning-enabled large language models (LLMs) excel in logical tasks, yet their utility for evaluating natural language generation remains unexplored. This study systematically compares reasoning LLMs with non-reasoning counterparts across machine translation and text summarization evaluation tasks. We evaluate eight models spanning state-of-the-art reasoning models (DeepSeek-R1, OpenAI o3), their distilled variants (8B-70B parameters), and equivalent non-reasoning LLMs. Experiments on WMT23 and SummEval benchmarks reveal architecture and task-dependent benefits: OpenAI o3-mini models show im"},"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":"2504.08120","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-10T20:39:18Z","cross_cats_sorted":[],"title_canon_sha256":"c649e284bf4679c16522599551a098c99ee9b685e2068ea230d212561816b22d","abstract_canon_sha256":"54f54c3bf1ae7bdb9274bcf23b4002fbd0146075c80f73b1fc57b122d0de022a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:12:30.629259Z","signature_b64":"3YA6vHEAjHCeJ9BBjbSCkCXfrtheyDusauX3miGHyDCWztE+plfkKRrKlJ0UETLh/pMn9yZQSFnWlc6uz8U1BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ebb6a7c6b090f67aea98bd1d56fbab4b470fd73abee2a4da3e60587dc245162b","last_reissued_at":"2026-07-05T11:12:30.628723Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:12:30.628723Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DeepSeek-R1 vs. o3-mini: How Well can Reasoning LLMs Evaluate MT and Summarization?","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Christian Greisinger, Christoph Leiter, Daniil Larionov, Ran Zhang, Sotaro Takeshita, Steffen Eger, Yanran Chen, Zhipin Wang","submitted_at":"2025-04-10T20:39:18Z","abstract_excerpt":"Reasoning-enabled large language models (LLMs) excel in logical tasks, yet their utility for evaluating natural language generation remains unexplored. This study systematically compares reasoning LLMs with non-reasoning counterparts across machine translation and text summarization evaluation tasks. We evaluate eight models spanning state-of-the-art reasoning models (DeepSeek-R1, OpenAI o3), their distilled variants (8B-70B parameters), and equivalent non-reasoning LLMs. Experiments on WMT23 and SummEval benchmarks reveal architecture and task-dependent benefits: OpenAI o3-mini models show im"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.08120","kind":"arxiv","version":3},"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/2504.08120/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":"2504.08120","created_at":"2026-07-05T11:12:30.628791+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.08120v3","created_at":"2026-07-05T11:12:30.628791+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.08120","created_at":"2026-07-05T11:12:30.628791+00:00"},{"alias_kind":"pith_short_12","alias_value":"5O3KPRVQSD3H","created_at":"2026-07-05T11:12:30.628791+00:00"},{"alias_kind":"pith_short_16","alias_value":"5O3KPRVQSD3HV2UY","created_at":"2026-07-05T11:12:30.628791+00:00"},{"alias_kind":"pith_short_8","alias_value":"5O3KPRVQ","created_at":"2026-07-05T11:12:30.628791+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.19144","citing_title":"ReflectMT: Internalizing Reflection for Efficient and High-Quality Machine Translation","ref_index":19,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5O3KPRVQSD3HV2UYXUOVN65LJN","json":"https://pith.science/pith/5O3KPRVQSD3HV2UYXUOVN65LJN.json","graph_json":"https://pith.science/api/pith-number/5O3KPRVQSD3HV2UYXUOVN65LJN/graph.json","events_json":"https://pith.science/api/pith-number/5O3KPRVQSD3HV2UYXUOVN65LJN/events.json","paper":"https://pith.science/paper/5O3KPRVQ"},"agent_actions":{"view_html":"https://pith.science/pith/5O3KPRVQSD3HV2UYXUOVN65LJN","download_json":"https://pith.science/pith/5O3KPRVQSD3HV2UYXUOVN65LJN.json","view_paper":"https://pith.science/paper/5O3KPRVQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.08120&json=true","fetch_graph":"https://pith.science/api/pith-number/5O3KPRVQSD3HV2UYXUOVN65LJN/graph.json","fetch_events":"https://pith.science/api/pith-number/5O3KPRVQSD3HV2UYXUOVN65LJN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5O3KPRVQSD3HV2UYXUOVN65LJN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5O3KPRVQSD3HV2UYXUOVN65LJN/action/storage_attestation","attest_author":"https://pith.science/pith/5O3KPRVQSD3HV2UYXUOVN65LJN/action/author_attestation","sign_citation":"https://pith.science/pith/5O3KPRVQSD3HV2UYXUOVN65LJN/action/citation_signature","submit_replication":"https://pith.science/pith/5O3KPRVQSD3HV2UYXUOVN65LJN/action/replication_record"}},"created_at":"2026-07-05T11:12:30.628791+00:00","updated_at":"2026-07-05T11:12:30.628791+00:00"}