{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:O4SBU4ONHXPX7XGALMJDGNY62W","short_pith_number":"pith:O4SBU4ON","canonical_record":{"source":{"id":"2505.10717","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-15T21:40:21Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"c3b5d7f9c715ba72cc099af7ad90172d71fdb05ace2282f49207b2c93cb65da8","abstract_canon_sha256":"fb31f45aa8ed8452f5e85b9bb52e4f07507cfe017808f921a3357cf5e1b8aac1"},"schema_version":"1.0"},"canonical_sha256":"77241a71cd3ddf7fdcc05b1233371ed5b834e958540b02db3b2bac8152ee038a","source":{"kind":"arxiv","id":"2505.10717","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.10717","created_at":"2026-07-05T11:59:14Z"},{"alias_kind":"arxiv_version","alias_value":"2505.10717v2","created_at":"2026-07-05T11:59:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.10717","created_at":"2026-07-05T11:59:14Z"},{"alias_kind":"pith_short_12","alias_value":"O4SBU4ONHXPX","created_at":"2026-07-05T11:59:14Z"},{"alias_kind":"pith_short_16","alias_value":"O4SBU4ONHXPX7XGA","created_at":"2026-07-05T11:59:14Z"},{"alias_kind":"pith_short_8","alias_value":"O4SBU4ON","created_at":"2026-07-05T11:59:14Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:O4SBU4ONHXPX7XGALMJDGNY62W","target":"record","payload":{"canonical_record":{"source":{"id":"2505.10717","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-15T21:40:21Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"c3b5d7f9c715ba72cc099af7ad90172d71fdb05ace2282f49207b2c93cb65da8","abstract_canon_sha256":"fb31f45aa8ed8452f5e85b9bb52e4f07507cfe017808f921a3357cf5e1b8aac1"},"schema_version":"1.0"},"canonical_sha256":"77241a71cd3ddf7fdcc05b1233371ed5b834e958540b02db3b2bac8152ee038a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:59:14.565567Z","signature_b64":"z7VJ8DUyw7cUPt0NI0j79aetQzYcDfV9uKHpIWvg2x+jn1FOxOWLHMHQypyyivMkyE9V81QTS5tkfjs9SULnBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"77241a71cd3ddf7fdcc05b1233371ed5b834e958540b02db3b2bac8152ee038a","last_reissued_at":"2026-07-05T11:59:14.565105Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:59:14.565105Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.10717","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:59:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qnAXLTfU0vmUNzWdkGliqOdQQu6Bty5krlMQzH1XlNyVgvx67GxpHn+nSyqzoRjpbmz/Vhw7BpwCIus99hdUCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T12:42:50.989705Z"},"content_sha256":"7aae42708b4c2633c62d10710b11006439a9d7c60e4d66c89c03dec2024d23d1","schema_version":"1.0","event_id":"sha256:7aae42708b4c2633c62d10710b11006439a9d7c60e4d66c89c03dec2024d23d1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:O4SBU4ONHXPX7XGALMJDGNY62W","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"A Modular Approach for Clinical SLMs Driven by Synthetic Data with Pre-Instruction Tuning, Model Merging, and Clinical-Tasks Alignment","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Alessandro Sordoni, Amin Dada, Asma Ben Abacha, Fran\\c{c}ois Beaulieu, Jean-Michel Attendu, Jean-Philippe Corbeil, Jens Kleesiek, Lucas Caccia, Paul Vozila, Thomas Lin","submitted_at":"2025-05-15T21:40:21Z","abstract_excerpt":"High computation costs and latency of large language models such as GPT-4 have limited their deployment in clinical settings. Small language models (SLMs) offer a cost-effective alternative, but their limited capacity requires biomedical domain adaptation, which remains challenging. An additional bottleneck is the unavailability and high sensitivity of clinical data. To address these challenges, we propose a novel framework for adapting SLMs into high-performing clinical models. We introduce the MediPhi collection of 3.8B-parameter SLMs developed with our novel framework: pre-instruction tunin"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.10717","kind":"arxiv","version":2},"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/2505.10717/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":"2026-07-05T11:59:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Du3nLDvZqtmHkIVK3pyfX0zNMSTBC6D4Rq5AFOz6fiMGCB7os7tWqewxvI5+JAKTOyZuhIjgWgAl1nfXI6BKDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T12:42:50.990077Z"},"content_sha256":"c7a2ee03ebc243469c9e56f541d4865a448efaefdc0eb39149a31b4f8d2d19e4","schema_version":"1.0","event_id":"sha256:c7a2ee03ebc243469c9e56f541d4865a448efaefdc0eb39149a31b4f8d2d19e4"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/O4SBU4ONHXPX7XGALMJDGNY62W/bundle.json","state_url":"https://pith.science/pith/O4SBU4ONHXPX7XGALMJDGNY62W/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/O4SBU4ONHXPX7XGALMJDGNY62W/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":"2026-07-07T12:42:50Z","links":{"resolver":"https://pith.science/pith/O4SBU4ONHXPX7XGALMJDGNY62W","bundle":"https://pith.science/pith/O4SBU4ONHXPX7XGALMJDGNY62W/bundle.json","state":"https://pith.science/pith/O4SBU4ONHXPX7XGALMJDGNY62W/state.json","well_known_bundle":"https://pith.science/.well-known/pith/O4SBU4ONHXPX7XGALMJDGNY62W/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:O4SBU4ONHXPX7XGALMJDGNY62W","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":"fb31f45aa8ed8452f5e85b9bb52e4f07507cfe017808f921a3357cf5e1b8aac1","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-15T21:40:21Z","title_canon_sha256":"c3b5d7f9c715ba72cc099af7ad90172d71fdb05ace2282f49207b2c93cb65da8"},"schema_version":"1.0","source":{"id":"2505.10717","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.10717","created_at":"2026-07-05T11:59:14Z"},{"alias_kind":"arxiv_version","alias_value":"2505.10717v2","created_at":"2026-07-05T11:59:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.10717","created_at":"2026-07-05T11:59:14Z"},{"alias_kind":"pith_short_12","alias_value":"O4SBU4ONHXPX","created_at":"2026-07-05T11:59:14Z"},{"alias_kind":"pith_short_16","alias_value":"O4SBU4ONHXPX7XGA","created_at":"2026-07-05T11:59:14Z"},{"alias_kind":"pith_short_8","alias_value":"O4SBU4ON","created_at":"2026-07-05T11:59:14Z"}],"graph_snapshots":[{"event_id":"sha256:c7a2ee03ebc243469c9e56f541d4865a448efaefdc0eb39149a31b4f8d2d19e4","target":"graph","created_at":"2026-07-05T11:59:14Z","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/2505.10717/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"High computation costs and latency of large language models such as GPT-4 have limited their deployment in clinical settings. Small language models (SLMs) offer a cost-effective alternative, but their limited capacity requires biomedical domain adaptation, which remains challenging. An additional bottleneck is the unavailability and high sensitivity of clinical data. To address these challenges, we propose a novel framework for adapting SLMs into high-performing clinical models. We introduce the MediPhi collection of 3.8B-parameter SLMs developed with our novel framework: pre-instruction tunin","authors_text":"Alessandro Sordoni, Amin Dada, Asma Ben Abacha, Fran\\c{c}ois Beaulieu, Jean-Michel Attendu, Jean-Philippe Corbeil, Jens Kleesiek, Lucas Caccia, Paul Vozila, Thomas Lin","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-15T21:40:21Z","title":"A Modular Approach for Clinical SLMs Driven by Synthetic Data with Pre-Instruction Tuning, Model Merging, and Clinical-Tasks Alignment"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.10717","kind":"arxiv","version":2},"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":"sha256:7aae42708b4c2633c62d10710b11006439a9d7c60e4d66c89c03dec2024d23d1","target":"record","created_at":"2026-07-05T11:59:14Z","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":"fb31f45aa8ed8452f5e85b9bb52e4f07507cfe017808f921a3357cf5e1b8aac1","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-15T21:40:21Z","title_canon_sha256":"c3b5d7f9c715ba72cc099af7ad90172d71fdb05ace2282f49207b2c93cb65da8"},"schema_version":"1.0","source":{"id":"2505.10717","kind":"arxiv","version":2}},"canonical_sha256":"77241a71cd3ddf7fdcc05b1233371ed5b834e958540b02db3b2bac8152ee038a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"77241a71cd3ddf7fdcc05b1233371ed5b834e958540b02db3b2bac8152ee038a","first_computed_at":"2026-07-05T11:59:14.565105Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:59:14.565105Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"z7VJ8DUyw7cUPt0NI0j79aetQzYcDfV9uKHpIWvg2x+jn1FOxOWLHMHQypyyivMkyE9V81QTS5tkfjs9SULnBA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:59:14.565567Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.10717","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7aae42708b4c2633c62d10710b11006439a9d7c60e4d66c89c03dec2024d23d1","sha256:c7a2ee03ebc243469c9e56f541d4865a448efaefdc0eb39149a31b4f8d2d19e4"],"state_sha256":"0b74e1117b011347e1a1033e37f482a19266bd63cdafc2f53f83d5211dd39aa1"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OAOonPess362UXSVgPeOZcwSxvCAg0nYvVfOHPpQSjT2YbuQMaRlOENYczZlzvdxXISTGi/BkztXbY6DOQrEBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T12:42:50.992092Z","bundle_sha256":"fa982aae8e24384c084bb9f6aecb5ae3b6d1440430168104dfa165191d2e9cce"}}