{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:XE53YHE5TLJRR33643XZTDBEZK","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"33511f73de544b6b2736cb03b464fd39d534fabba3e401a825d7736848cad633","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-27T11:16:19Z","title_canon_sha256":"e7716cf6aac74499775ded9733930d93fbd6a05190c5befe67c2688d51672a10"},"schema_version":"1.0","source":{"id":"2605.28315","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.28315","created_at":"2026-05-28T01:05:06Z"},{"alias_kind":"arxiv_version","alias_value":"2605.28315v1","created_at":"2026-05-28T01:05:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.28315","created_at":"2026-05-28T01:05:06Z"},{"alias_kind":"pith_short_12","alias_value":"XE53YHE5TLJR","created_at":"2026-05-28T01:05:06Z"},{"alias_kind":"pith_short_16","alias_value":"XE53YHE5TLJRR336","created_at":"2026-05-28T01:05:06Z"},{"alias_kind":"pith_short_8","alias_value":"XE53YHE5","created_at":"2026-05-28T01:05:06Z"}],"graph_snapshots":[{"event_id":"sha256:3272a3e4379fc8a0e6b80e358cc37ff3b3b6047774d13bad2e23567c1b5abdd0","target":"graph","created_at":"2026-05-28T01:05:06Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2605.28315/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"General-purpose machine translation benchmarks such as FLORES-200 have reached a saturation regime on Chinese-English pairs, where modern large language models cluster within a narrow band of high scores. Across 22 systems, FLORES-200 zh-en GEMBA scores fall in a 7.87-point range with a standard deviation of 2.29, which compresses the separation between systems on knowledge-intensive domains such as finance, healthcare, law, and science and technology. We introduce HardMTBench, a difficulty-aware diagnostic benchmark for bidirectional Chinese-English domain translation. HardMTBench covers 12 d","authors_text":"Mao Zheng, Mingyang Song, Tianxiang Fei, Zheng Li","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-27T11:16:19Z","title":"HardMTBench: Stress-Testing Chinese-English Translation on Knowledge-Intensive Domains"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.28315","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:5e4e68601b031c7dcf65598053360c5b037a5d695d2b9734573822059748e9ff","target":"record","created_at":"2026-05-28T01:05:06Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"33511f73de544b6b2736cb03b464fd39d534fabba3e401a825d7736848cad633","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-27T11:16:19Z","title_canon_sha256":"e7716cf6aac74499775ded9733930d93fbd6a05190c5befe67c2688d51672a10"},"schema_version":"1.0","source":{"id":"2605.28315","kind":"arxiv","version":1}},"canonical_sha256":"b93bbc1c9d9ad318ef7ee6ef998c24caa1c35302d2e115042f2dfa1115a23187","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b93bbc1c9d9ad318ef7ee6ef998c24caa1c35302d2e115042f2dfa1115a23187","first_computed_at":"2026-05-28T01:05:06.177589Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-28T01:05:06.177589Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"KyhtS+YsLbCPfYKDn7pgczv2wdcwFCCMOSgubSUIdxJklS0MhnuvBecInK0gXiFdifR3xRbkMfc9EDkBpewECA==","signature_status":"signed_v1","signed_at":"2026-05-28T01:05:06.178070Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.28315","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5e4e68601b031c7dcf65598053360c5b037a5d695d2b9734573822059748e9ff","sha256:3272a3e4379fc8a0e6b80e358cc37ff3b3b6047774d13bad2e23567c1b5abdd0"],"state_sha256":"2e63e1effb78dfdfa5565ba235951d88e8d126c0762987541cb556484d3e2cc7"}