{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:ZKRHLRWR6YQB37FFPTR5EYWT4Q","short_pith_number":"pith:ZKRHLRWR","canonical_record":{"source":{"id":"2605.19357","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-19T04:41:43Z","cross_cats_sorted":[],"title_canon_sha256":"c786d2477daf243f602abd5f482e334af58032ff219414a6c90068e0cceb2aa0","abstract_canon_sha256":"8006ef1a2d87a8ee050eb7a8631a4bdf7c7d08a0ee46185b3a4680c98908ff23"},"schema_version":"1.0"},"canonical_sha256":"caa275c6d1f6201dfca57ce3d262d3e40963009f405fd0350e458a67d99fe19c","source":{"kind":"arxiv","id":"2605.19357","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.19357","created_at":"2026-05-20T01:05:41Z"},{"alias_kind":"arxiv_version","alias_value":"2605.19357v1","created_at":"2026-05-20T01:05:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.19357","created_at":"2026-05-20T01:05:41Z"},{"alias_kind":"pith_short_12","alias_value":"ZKRHLRWR6YQB","created_at":"2026-05-20T01:05:41Z"},{"alias_kind":"pith_short_16","alias_value":"ZKRHLRWR6YQB37FF","created_at":"2026-05-20T01:05:41Z"},{"alias_kind":"pith_short_8","alias_value":"ZKRHLRWR","created_at":"2026-05-20T01:05:41Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:ZKRHLRWR6YQB37FFPTR5EYWT4Q","target":"record","payload":{"canonical_record":{"source":{"id":"2605.19357","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-19T04:41:43Z","cross_cats_sorted":[],"title_canon_sha256":"c786d2477daf243f602abd5f482e334af58032ff219414a6c90068e0cceb2aa0","abstract_canon_sha256":"8006ef1a2d87a8ee050eb7a8631a4bdf7c7d08a0ee46185b3a4680c98908ff23"},"schema_version":"1.0"},"canonical_sha256":"caa275c6d1f6201dfca57ce3d262d3e40963009f405fd0350e458a67d99fe19c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T01:05:41.351970Z","signature_b64":"vV4BjLzmp0HssD1ZhKuUM9pZRh7v63fzfQl8XtGb9ZOqEmjADUt5ey3sTnQX/KDNczcw23oiwl8v+pom+Mq9Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"caa275c6d1f6201dfca57ce3d262d3e40963009f405fd0350e458a67d99fe19c","last_reissued_at":"2026-05-20T01:05:41.351436Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T01:05:41.351436Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.19357","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":"2026-05-20T01:05:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"29pqMo5C15jEg+ORqMfMviZJcbfgKmGEvZSCESFgA5LH/x7kfwz3Rn8/SsIoxkX7Hj+mHARCDE0Q8N3Exhp3Cg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-29T17:31:39.506532Z"},"content_sha256":"5dc421afe4ca8cdc7d364f8a6f53526168d6cd196f2f910bd1e32ab07022e494","schema_version":"1.0","event_id":"sha256:5dc421afe4ca8cdc7d364f8a6f53526168d6cd196f2f910bd1e32ab07022e494"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:ZKRHLRWR6YQB37FFPTR5EYWT4Q","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"SciCustom: A Framework for Custom Evaluation of Scientific Capabilities in Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Beier Xiao, Bin Feng, Bohan Wu, Haoran Li, Junwei Yang, Junyu Luo, Kaili Liu, Ming Zhang, Philip S. Yu, Qi Shi, Shufang Xie, Weizhi Zhang, Xiao Luo, Ye Yuan, Yingce Xia, Yiyang Gu, Zequn Liu, Zhiping Xiao","submitted_at":"2026-05-19T04:41:43Z","abstract_excerpt":"Large language models (LLMs) are increasingly applied to scientific research, yet existing evaluations often fail to reflect the fine-grained capabilities required in practice. Most benchmarks are manually curated or domain-generic, limiting scalability and alignment with real scientific use cases. In this paper, we propose a new framework named SciCustom to address the problem. It enables the custom construction of benchmarks from large-scale scientific data to evaluate application-specific scientific capabilities in LLMs. SciCustom first organizes scientific knowledge into ontology-grounded "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.19357","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/2605.19357/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":"2026-05-20T01:05:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"x7Sxr/r16SZqIeWXCZr269CaS9mn94XI8EVXe27mNcn5urEiqvf99cChmgKQIgzbgcTLQ3SaWPdJbvBAViTwAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-29T17:31:39.507241Z"},"content_sha256":"41c43ec40651600856c9b5e0dc4856d2848130cdb39843f798dab1191aea18f5","schema_version":"1.0","event_id":"sha256:41c43ec40651600856c9b5e0dc4856d2848130cdb39843f798dab1191aea18f5"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ZKRHLRWR6YQB37FFPTR5EYWT4Q/bundle.json","state_url":"https://pith.science/pith/ZKRHLRWR6YQB37FFPTR5EYWT4Q/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ZKRHLRWR6YQB37FFPTR5EYWT4Q/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":"2026-05-29T17:31:39Z","links":{"resolver":"https://pith.science/pith/ZKRHLRWR6YQB37FFPTR5EYWT4Q","bundle":"https://pith.science/pith/ZKRHLRWR6YQB37FFPTR5EYWT4Q/bundle.json","state":"https://pith.science/pith/ZKRHLRWR6YQB37FFPTR5EYWT4Q/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ZKRHLRWR6YQB37FFPTR5EYWT4Q/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:ZKRHLRWR6YQB37FFPTR5EYWT4Q","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":"8006ef1a2d87a8ee050eb7a8631a4bdf7c7d08a0ee46185b3a4680c98908ff23","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-19T04:41:43Z","title_canon_sha256":"c786d2477daf243f602abd5f482e334af58032ff219414a6c90068e0cceb2aa0"},"schema_version":"1.0","source":{"id":"2605.19357","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.19357","created_at":"2026-05-20T01:05:41Z"},{"alias_kind":"arxiv_version","alias_value":"2605.19357v1","created_at":"2026-05-20T01:05:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.19357","created_at":"2026-05-20T01:05:41Z"},{"alias_kind":"pith_short_12","alias_value":"ZKRHLRWR6YQB","created_at":"2026-05-20T01:05:41Z"},{"alias_kind":"pith_short_16","alias_value":"ZKRHLRWR6YQB37FF","created_at":"2026-05-20T01:05:41Z"},{"alias_kind":"pith_short_8","alias_value":"ZKRHLRWR","created_at":"2026-05-20T01:05:41Z"}],"graph_snapshots":[{"event_id":"sha256:41c43ec40651600856c9b5e0dc4856d2848130cdb39843f798dab1191aea18f5","target":"graph","created_at":"2026-05-20T01:05:41Z","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/2605.19357/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) are increasingly applied to scientific research, yet existing evaluations often fail to reflect the fine-grained capabilities required in practice. Most benchmarks are manually curated or domain-generic, limiting scalability and alignment with real scientific use cases. In this paper, we propose a new framework named SciCustom to address the problem. It enables the custom construction of benchmarks from large-scale scientific data to evaluate application-specific scientific capabilities in LLMs. SciCustom first organizes scientific knowledge into ontology-grounded ","authors_text":"Beier Xiao, Bin Feng, Bohan Wu, Haoran Li, Junwei Yang, Junyu Luo, Kaili Liu, Ming Zhang, Philip S. Yu, Qi Shi, Shufang Xie, Weizhi Zhang, Xiao Luo, Ye Yuan, Yingce Xia, Yiyang Gu, Zequn Liu, Zhiping Xiao","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-19T04:41:43Z","title":"SciCustom: A Framework for Custom Evaluation of Scientific Capabilities in Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.19357","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":"sha256:5dc421afe4ca8cdc7d364f8a6f53526168d6cd196f2f910bd1e32ab07022e494","target":"record","created_at":"2026-05-20T01:05:41Z","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":"8006ef1a2d87a8ee050eb7a8631a4bdf7c7d08a0ee46185b3a4680c98908ff23","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-19T04:41:43Z","title_canon_sha256":"c786d2477daf243f602abd5f482e334af58032ff219414a6c90068e0cceb2aa0"},"schema_version":"1.0","source":{"id":"2605.19357","kind":"arxiv","version":1}},"canonical_sha256":"caa275c6d1f6201dfca57ce3d262d3e40963009f405fd0350e458a67d99fe19c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"caa275c6d1f6201dfca57ce3d262d3e40963009f405fd0350e458a67d99fe19c","first_computed_at":"2026-05-20T01:05:41.351436Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T01:05:41.351436Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"vV4BjLzmp0HssD1ZhKuUM9pZRh7v63fzfQl8XtGb9ZOqEmjADUt5ey3sTnQX/KDNczcw23oiwl8v+pom+Mq9Aw==","signature_status":"signed_v1","signed_at":"2026-05-20T01:05:41.351970Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.19357","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5dc421afe4ca8cdc7d364f8a6f53526168d6cd196f2f910bd1e32ab07022e494","sha256:41c43ec40651600856c9b5e0dc4856d2848130cdb39843f798dab1191aea18f5"],"state_sha256":"f594923fff776536bbd99eabf1f223ddc8c5f2468c9773557a1d215f15611d2d"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TlGYx1CzweHTYqhqmNhXbIdxpYrx33pkZtJH+oiurx6hHu2x+p6KF5e31ZDAq6ybgvd1R1Dt4Vm68RbW7Ec7Ag==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-29T17:31:39.511434Z","bundle_sha256":"d5a0784fc5b22cf7066d06343ebe06c3903f60050b2a296c9227e67783213014"}}