{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:2XIJ3KSGMHJTJFFGYRC66E5AQJ","short_pith_number":"pith:2XIJ3KSG","canonical_record":{"source":{"id":"2506.12909","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-15T16:57:14Z","cross_cats_sorted":[],"title_canon_sha256":"02997c9d5ede2f4ae8cf309653d2f8381b248006ead890570d4853536baded3a","abstract_canon_sha256":"5260e6f8f11689f530f2e38be44fcfd554dd7afe7009a2f5beaa67818cf8c1db"},"schema_version":"1.0"},"canonical_sha256":"d5d09daa4661d33494a6c445ef13a0827d92ea16503dc812a060e33fd08a95e2","source":{"kind":"arxiv","id":"2506.12909","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.12909","created_at":"2026-07-05T11:22:02Z"},{"alias_kind":"arxiv_version","alias_value":"2506.12909v1","created_at":"2026-07-05T11:22:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.12909","created_at":"2026-07-05T11:22:02Z"},{"alias_kind":"pith_short_12","alias_value":"2XIJ3KSGMHJT","created_at":"2026-07-05T11:22:02Z"},{"alias_kind":"pith_short_16","alias_value":"2XIJ3KSGMHJTJFFG","created_at":"2026-07-05T11:22:02Z"},{"alias_kind":"pith_short_8","alias_value":"2XIJ3KSG","created_at":"2026-07-05T11:22:02Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:2XIJ3KSGMHJTJFFGYRC66E5AQJ","target":"record","payload":{"canonical_record":{"source":{"id":"2506.12909","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-15T16:57:14Z","cross_cats_sorted":[],"title_canon_sha256":"02997c9d5ede2f4ae8cf309653d2f8381b248006ead890570d4853536baded3a","abstract_canon_sha256":"5260e6f8f11689f530f2e38be44fcfd554dd7afe7009a2f5beaa67818cf8c1db"},"schema_version":"1.0"},"canonical_sha256":"d5d09daa4661d33494a6c445ef13a0827d92ea16503dc812a060e33fd08a95e2","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:22:02.187857Z","signature_b64":"yaFL4gnP4uBXA06mn3smBTrC4mEeO9MeSrV0HlhTYz9WhWQ2YBXr/zm1FEexHRR0R/KCmKrHI3myInsYdzRrCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d5d09daa4661d33494a6c445ef13a0827d92ea16503dc812a060e33fd08a95e2","last_reissued_at":"2026-07-05T11:22:02.187428Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:22:02.187428Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.12909","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:22:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"bjGjmIADDDGM0Jk83zApOUcSuvCbKPvD0LN+kKHmG8E1IwmbX35/Ktp39iP3s6XMRiPmL+jIABzkwlfi7M1HAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-22T15:06:13.442447Z"},"content_sha256":"4fb483779131bd6209adea140d54c4b3f5a519ebf472f568a660aef6593164d4","schema_version":"1.0","event_id":"sha256:4fb483779131bd6209adea140d54c4b3f5a519ebf472f568a660aef6593164d4"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:2XIJ3KSGMHJTJFFGYRC66E5AQJ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"SciDA: Scientific Dynamic Assessor of LLMs","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chen Dun, Ge Yan, Ge Zhang, Hongwan Gao, Junting Zhou, Leqi Wang, Qichao Wang, Renjie Zheng, Tingjia Miao, Tong Yang, Wenhao Huang, Yanqin Wang, Yitao Liang, Yiyan Liao, Yuansong Zeng, Yucheng Xia, Yunjie Huang, Zhoufutu Wen","submitted_at":"2025-06-15T16:57:14Z","abstract_excerpt":"Advancement in Large Language Models (LLMs) reasoning capabilities enables them to solve scientific problems with enhanced efficacy. Thereby, a high-quality benchmark for comprehensive and appropriate assessment holds significance, while existing ones either confront the risk of data contamination or lack involved disciplines. To be specific, due to the data source overlap of LLMs training and static benchmark, the keys or number pattern of answers inadvertently memorized (i.e. data contamination), leading to systematic overestimation of their reasoning capabilities, especially numerical reaso"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.12909","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/2506.12909/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:22:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"pBnQ0oMOjqWgBx0IO2RZMRMA6FUxlpeCBDXJ04jYsYvQXGHkizUiwGA8rxcPjBbK4KC0oT7qmthizRChDe2cDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-22T15:06:13.442942Z"},"content_sha256":"96cf41a33a915abed889581eb5b3bebb2589d8840b32ed1cf9e6d19db37b2f48","schema_version":"1.0","event_id":"sha256:96cf41a33a915abed889581eb5b3bebb2589d8840b32ed1cf9e6d19db37b2f48"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/2XIJ3KSGMHJTJFFGYRC66E5AQJ/bundle.json","state_url":"https://pith.science/pith/2XIJ3KSGMHJTJFFGYRC66E5AQJ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/2XIJ3KSGMHJTJFFGYRC66E5AQJ/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-22T15:06:13Z","links":{"resolver":"https://pith.science/pith/2XIJ3KSGMHJTJFFGYRC66E5AQJ","bundle":"https://pith.science/pith/2XIJ3KSGMHJTJFFGYRC66E5AQJ/bundle.json","state":"https://pith.science/pith/2XIJ3KSGMHJTJFFGYRC66E5AQJ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/2XIJ3KSGMHJTJFFGYRC66E5AQJ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:2XIJ3KSGMHJTJFFGYRC66E5AQJ","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":"5260e6f8f11689f530f2e38be44fcfd554dd7afe7009a2f5beaa67818cf8c1db","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-15T16:57:14Z","title_canon_sha256":"02997c9d5ede2f4ae8cf309653d2f8381b248006ead890570d4853536baded3a"},"schema_version":"1.0","source":{"id":"2506.12909","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.12909","created_at":"2026-07-05T11:22:02Z"},{"alias_kind":"arxiv_version","alias_value":"2506.12909v1","created_at":"2026-07-05T11:22:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.12909","created_at":"2026-07-05T11:22:02Z"},{"alias_kind":"pith_short_12","alias_value":"2XIJ3KSGMHJT","created_at":"2026-07-05T11:22:02Z"},{"alias_kind":"pith_short_16","alias_value":"2XIJ3KSGMHJTJFFG","created_at":"2026-07-05T11:22:02Z"},{"alias_kind":"pith_short_8","alias_value":"2XIJ3KSG","created_at":"2026-07-05T11:22:02Z"}],"graph_snapshots":[{"event_id":"sha256:96cf41a33a915abed889581eb5b3bebb2589d8840b32ed1cf9e6d19db37b2f48","target":"graph","created_at":"2026-07-05T11:22:02Z","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/2506.12909/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Advancement in Large Language Models (LLMs) reasoning capabilities enables them to solve scientific problems with enhanced efficacy. Thereby, a high-quality benchmark for comprehensive and appropriate assessment holds significance, while existing ones either confront the risk of data contamination or lack involved disciplines. To be specific, due to the data source overlap of LLMs training and static benchmark, the keys or number pattern of answers inadvertently memorized (i.e. data contamination), leading to systematic overestimation of their reasoning capabilities, especially numerical reaso","authors_text":"Chen Dun, Ge Yan, Ge Zhang, Hongwan Gao, Junting Zhou, Leqi Wang, Qichao Wang, Renjie Zheng, Tingjia Miao, Tong Yang, Wenhao Huang, Yanqin Wang, Yitao Liang, Yiyan Liao, Yuansong Zeng, Yucheng Xia, Yunjie Huang, Zhoufutu Wen","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-15T16:57:14Z","title":"SciDA: Scientific Dynamic Assessor of LLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.12909","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":"sha256:4fb483779131bd6209adea140d54c4b3f5a519ebf472f568a660aef6593164d4","target":"record","created_at":"2026-07-05T11:22:02Z","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":"5260e6f8f11689f530f2e38be44fcfd554dd7afe7009a2f5beaa67818cf8c1db","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-15T16:57:14Z","title_canon_sha256":"02997c9d5ede2f4ae8cf309653d2f8381b248006ead890570d4853536baded3a"},"schema_version":"1.0","source":{"id":"2506.12909","kind":"arxiv","version":1}},"canonical_sha256":"d5d09daa4661d33494a6c445ef13a0827d92ea16503dc812a060e33fd08a95e2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d5d09daa4661d33494a6c445ef13a0827d92ea16503dc812a060e33fd08a95e2","first_computed_at":"2026-07-05T11:22:02.187428Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:22:02.187428Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"yaFL4gnP4uBXA06mn3smBTrC4mEeO9MeSrV0HlhTYz9WhWQ2YBXr/zm1FEexHRR0R/KCmKrHI3myInsYdzRrCw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:22:02.187857Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.12909","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4fb483779131bd6209adea140d54c4b3f5a519ebf472f568a660aef6593164d4","sha256:96cf41a33a915abed889581eb5b3bebb2589d8840b32ed1cf9e6d19db37b2f48"],"state_sha256":"0706a6365d264cf000a3180e12d787b54bb3061fb8614408e8d3bfe4034ca458"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"tia3RI+ltT8NLLqFC7QbDm1byn5WfUFsfO+ChuBKoOH94lbavATbdIAVvXB2AQws0GM3CedqbkFNj/BI24laAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-22T15:06:13.447449Z","bundle_sha256":"a5eebce1c33a253b1b7a116868ac6617a84f708abac9c77fe88fd61e072314fe"}}