{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:M7O7MGVLNYFCUKG3ZOIAY4QUCW","short_pith_number":"pith:M7O7MGVL","canonical_record":{"source":{"id":"2607.02140","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-02T13:16:31Z","cross_cats_sorted":[],"title_canon_sha256":"bf2ff6e97e830afc9c064d9b81dea761dfdae294d82e16e524cc85c417401db1","abstract_canon_sha256":"2e4b0223a4921f74124923ee6d84f3980ea1ae2144db610d90a4695cb20ccf46"},"schema_version":"1.0"},"canonical_sha256":"67ddf61aab6e0a2a28dbcb900c7214158d0417fcb2bb5a563d919077ad640e4d","source":{"kind":"arxiv","id":"2607.02140","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.02140","created_at":"2026-07-03T01:17:43Z"},{"alias_kind":"arxiv_version","alias_value":"2607.02140v1","created_at":"2026-07-03T01:17:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.02140","created_at":"2026-07-03T01:17:43Z"},{"alias_kind":"pith_short_12","alias_value":"M7O7MGVLNYFC","created_at":"2026-07-03T01:17:43Z"},{"alias_kind":"pith_short_16","alias_value":"M7O7MGVLNYFCUKG3","created_at":"2026-07-03T01:17:43Z"},{"alias_kind":"pith_short_8","alias_value":"M7O7MGVL","created_at":"2026-07-03T01:17:43Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:M7O7MGVLNYFCUKG3ZOIAY4QUCW","target":"record","payload":{"canonical_record":{"source":{"id":"2607.02140","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-02T13:16:31Z","cross_cats_sorted":[],"title_canon_sha256":"bf2ff6e97e830afc9c064d9b81dea761dfdae294d82e16e524cc85c417401db1","abstract_canon_sha256":"2e4b0223a4921f74124923ee6d84f3980ea1ae2144db610d90a4695cb20ccf46"},"schema_version":"1.0"},"canonical_sha256":"67ddf61aab6e0a2a28dbcb900c7214158d0417fcb2bb5a563d919077ad640e4d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-03T01:17:43.233766Z","signature_b64":"wcYN5y4f76if0aY4L6QYYR83+gByCF5QlxDYpuSLMVMPFS6Fj3zLKKfq0CZfgSl8y37QoFcmVRpg5gbCueAPDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"67ddf61aab6e0a2a28dbcb900c7214158d0417fcb2bb5a563d919077ad640e4d","last_reissued_at":"2026-07-03T01:17:43.233335Z","signature_status":"signed_v1","first_computed_at":"2026-07-03T01:17:43.233335Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2607.02140","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":"2026-07-03T01:17:43Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"WT0x+briFmQyUB3dHidGRurskYeujqzRl6ic3bZ11K52K9XZP+CfvzmqMcdw/O9Yhp3Z906i4VM6hfrCiAueAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-04T06:18:37.245208Z"},"content_sha256":"8eb07fb92b6f6390b53428c88ab82fded834aceb8b8441bcf077c4a3da41a5ae","schema_version":"1.0","event_id":"sha256:8eb07fb92b6f6390b53428c88ab82fded834aceb8b8441bcf077c4a3da41a5ae"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:M7O7MGVLNYFCUKG3ZOIAY4QUCW","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Probing Chemical Language Models: Effects of Pre-training and Fine-tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Anna Karnysheva, Dietrich Klakow, Ji-Ung Lee","submitted_at":"2026-07-02T13:16:31Z","abstract_excerpt":"Chemical language models (CLMs) are trained with linearized representations such as SMILES, yet it remains unclear which chemically meaningful substructures they encode. To foster a better understanding of CLMs, we conduct a systematic study and probe for 78 molecular substructures across eight pre-trained and six randomly initialized models. We furthermore study how fine-tuning on chemical downstream tasks affects the learned representations of molecular substructures. Our results show that pre-training generally improves molecular structure awareness of CLMs, particularly in the upper layers"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.02140","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/2607.02140/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":"2026-07-03T01:17:43Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SXLonvpgi67aWmuu7aLzlj3rvurhpYl4KmYyK3k2zgYqCMKs58/uFtG1lgWcwT8oErRdJZco82ISWK5ENzIJCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-04T06:18:37.245607Z"},"content_sha256":"d48fa454ba6e1e759084013157fd3ff49bc012f39c3ce6e4690938d6ea4958ba","schema_version":"1.0","event_id":"sha256:d48fa454ba6e1e759084013157fd3ff49bc012f39c3ce6e4690938d6ea4958ba"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/M7O7MGVLNYFCUKG3ZOIAY4QUCW/bundle.json","state_url":"https://pith.science/pith/M7O7MGVLNYFCUKG3ZOIAY4QUCW/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/M7O7MGVLNYFCUKG3ZOIAY4QUCW/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":"2026-07-04T06:18:37Z","links":{"resolver":"https://pith.science/pith/M7O7MGVLNYFCUKG3ZOIAY4QUCW","bundle":"https://pith.science/pith/M7O7MGVLNYFCUKG3ZOIAY4QUCW/bundle.json","state":"https://pith.science/pith/M7O7MGVLNYFCUKG3ZOIAY4QUCW/state.json","well_known_bundle":"https://pith.science/.well-known/pith/M7O7MGVLNYFCUKG3ZOIAY4QUCW/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:M7O7MGVLNYFCUKG3ZOIAY4QUCW","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":"2e4b0223a4921f74124923ee6d84f3980ea1ae2144db610d90a4695cb20ccf46","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-02T13:16:31Z","title_canon_sha256":"bf2ff6e97e830afc9c064d9b81dea761dfdae294d82e16e524cc85c417401db1"},"schema_version":"1.0","source":{"id":"2607.02140","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.02140","created_at":"2026-07-03T01:17:43Z"},{"alias_kind":"arxiv_version","alias_value":"2607.02140v1","created_at":"2026-07-03T01:17:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.02140","created_at":"2026-07-03T01:17:43Z"},{"alias_kind":"pith_short_12","alias_value":"M7O7MGVLNYFC","created_at":"2026-07-03T01:17:43Z"},{"alias_kind":"pith_short_16","alias_value":"M7O7MGVLNYFCUKG3","created_at":"2026-07-03T01:17:43Z"},{"alias_kind":"pith_short_8","alias_value":"M7O7MGVL","created_at":"2026-07-03T01:17:43Z"}],"graph_snapshots":[{"event_id":"sha256:d48fa454ba6e1e759084013157fd3ff49bc012f39c3ce6e4690938d6ea4958ba","target":"graph","created_at":"2026-07-03T01:17:43Z","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/2607.02140/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Chemical language models (CLMs) are trained with linearized representations such as SMILES, yet it remains unclear which chemically meaningful substructures they encode. To foster a better understanding of CLMs, we conduct a systematic study and probe for 78 molecular substructures across eight pre-trained and six randomly initialized models. We furthermore study how fine-tuning on chemical downstream tasks affects the learned representations of molecular substructures. Our results show that pre-training generally improves molecular structure awareness of CLMs, particularly in the upper layers","authors_text":"Anna Karnysheva, Dietrich Klakow, Ji-Ung Lee","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-02T13:16:31Z","title":"Probing Chemical Language Models: Effects of Pre-training and Fine-tuning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.02140","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":"sha256:8eb07fb92b6f6390b53428c88ab82fded834aceb8b8441bcf077c4a3da41a5ae","target":"record","created_at":"2026-07-03T01:17:43Z","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":"2e4b0223a4921f74124923ee6d84f3980ea1ae2144db610d90a4695cb20ccf46","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-02T13:16:31Z","title_canon_sha256":"bf2ff6e97e830afc9c064d9b81dea761dfdae294d82e16e524cc85c417401db1"},"schema_version":"1.0","source":{"id":"2607.02140","kind":"arxiv","version":1}},"canonical_sha256":"67ddf61aab6e0a2a28dbcb900c7214158d0417fcb2bb5a563d919077ad640e4d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"67ddf61aab6e0a2a28dbcb900c7214158d0417fcb2bb5a563d919077ad640e4d","first_computed_at":"2026-07-03T01:17:43.233335Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-03T01:17:43.233335Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"wcYN5y4f76if0aY4L6QYYR83+gByCF5QlxDYpuSLMVMPFS6Fj3zLKKfq0CZfgSl8y37QoFcmVRpg5gbCueAPDw==","signature_status":"signed_v1","signed_at":"2026-07-03T01:17:43.233766Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.02140","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:8eb07fb92b6f6390b53428c88ab82fded834aceb8b8441bcf077c4a3da41a5ae","sha256:d48fa454ba6e1e759084013157fd3ff49bc012f39c3ce6e4690938d6ea4958ba"],"state_sha256":"1d804a2a7cafb4dc8b5a4eb7c5272bd049ca7c385dca01beeac2884beab4838d"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0R1fuXUk0Ayo2hVShUymxOD0hUyWIHq+EoK19n4Bq3MtSBR4UPZ+YSQU7E+FHbJUJEyGAaUmzV3LMccXRbvHCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-04T06:18:37.247734Z","bundle_sha256":"ece1e10631555b3d34507cca85e6324f84f52e367121a0da738d41926d601be6"}}