{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:YS43HFSYFHPKG4LNEMFQT6WTJN","short_pith_number":"pith:YS43HFSY","canonical_record":{"source":{"id":"2606.19496","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-17T18:35:16Z","cross_cats_sorted":[],"title_canon_sha256":"5d9989ac7af66daabae9ef68155964cc7249abc615eca14773d2d64efe49343e","abstract_canon_sha256":"27d42dadd4f9033607b79429281487a7133c03ec956ce33ea6e27598a0ed1173"},"schema_version":"1.0"},"canonical_sha256":"c4b9b3965829dea3716d230b09fad34b658412abb0fddd0bdd4c38fa53c87485","source":{"kind":"arxiv","id":"2606.19496","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.19496","created_at":"2026-06-19T16:12:27Z"},{"alias_kind":"arxiv_version","alias_value":"2606.19496v1","created_at":"2026-06-19T16:12:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.19496","created_at":"2026-06-19T16:12:27Z"},{"alias_kind":"pith_short_12","alias_value":"YS43HFSYFHPK","created_at":"2026-06-19T16:12:27Z"},{"alias_kind":"pith_short_16","alias_value":"YS43HFSYFHPKG4LN","created_at":"2026-06-19T16:12:27Z"},{"alias_kind":"pith_short_8","alias_value":"YS43HFSY","created_at":"2026-06-19T16:12:27Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:YS43HFSYFHPKG4LNEMFQT6WTJN","target":"record","payload":{"canonical_record":{"source":{"id":"2606.19496","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-17T18:35:16Z","cross_cats_sorted":[],"title_canon_sha256":"5d9989ac7af66daabae9ef68155964cc7249abc615eca14773d2d64efe49343e","abstract_canon_sha256":"27d42dadd4f9033607b79429281487a7133c03ec956ce33ea6e27598a0ed1173"},"schema_version":"1.0"},"canonical_sha256":"c4b9b3965829dea3716d230b09fad34b658412abb0fddd0bdd4c38fa53c87485","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-19T16:12:27.247142Z","signature_b64":"BehQutSjlOPkKFIiCiIFeAo8x6uPiYtJrZdbMPV/AnugsZCy/ZlOE0pFYHrnNMc6HZHGYyG0OcA2c7qwi7KEBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c4b9b3965829dea3716d230b09fad34b658412abb0fddd0bdd4c38fa53c87485","last_reissued_at":"2026-06-19T16:12:27.246797Z","signature_status":"signed_v1","first_computed_at":"2026-06-19T16:12:27.246797Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2606.19496","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":"2026-06-19T16:12:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3FjNq5Uav2/v4QcrxOJeBz8bP0S7fGaf51ge8GrC5vOtLDNXOiofz+ORLFUUy2yyERm6vZmlATHAErqatyeaDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-30T16:18:59.409199Z"},"content_sha256":"e1c036a1ac2b17a4defc1df9d3fc7daf0747343b82513ddae24e73aaa11f3c33","schema_version":"1.0","event_id":"sha256:e1c036a1ac2b17a4defc1df9d3fc7daf0747343b82513ddae24e73aaa11f3c33"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:YS43HFSYFHPKG4LNEMFQT6WTJN","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Calibrating Generative Models to Feature Distributions with MMD Finetuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Brian L. Trippe, Nathaniel L. Diamant","submitted_at":"2026-06-17T18:35:16Z","abstract_excerpt":"Generative models can produce individually plausible samples while deviating substantially from a target set in the distribution of key features. For example, a model pretrained on broad drug-like chemical space may generate molecules whose molecular features differ from those of a therapeutic class of interest, such as known antibiotics. Correcting such distributional miscalibration is challenging: direct finetuning on the target set can overfit and does not control which features are matched. To fill this gap, we introduce kernel Calibrating Generative Models (kCGM). kCGM minimizes a maximum"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.19496","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/2606.19496/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":"2026-06-19T16:12:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9Dx1aMkpTS3KK0UvGFPiMKKdW4bBY7IdIZ4lwgqgj3wo/M3ndHm3rdOxIOkBg+B3bbMaarkvH4k2vU8pFVuNDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-30T16:18:59.409583Z"},"content_sha256":"4ac0ab25278ce5035437990c695966b12c9e9729d3f3cb4d58141307e72dfffd","schema_version":"1.0","event_id":"sha256:4ac0ab25278ce5035437990c695966b12c9e9729d3f3cb4d58141307e72dfffd"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/YS43HFSYFHPKG4LNEMFQT6WTJN/bundle.json","state_url":"https://pith.science/pith/YS43HFSYFHPKG4LNEMFQT6WTJN/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/YS43HFSYFHPKG4LNEMFQT6WTJN/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":"2026-06-30T16:18:59Z","links":{"resolver":"https://pith.science/pith/YS43HFSYFHPKG4LNEMFQT6WTJN","bundle":"https://pith.science/pith/YS43HFSYFHPKG4LNEMFQT6WTJN/bundle.json","state":"https://pith.science/pith/YS43HFSYFHPKG4LNEMFQT6WTJN/state.json","well_known_bundle":"https://pith.science/.well-known/pith/YS43HFSYFHPKG4LNEMFQT6WTJN/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:YS43HFSYFHPKG4LNEMFQT6WTJN","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":"27d42dadd4f9033607b79429281487a7133c03ec956ce33ea6e27598a0ed1173","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-17T18:35:16Z","title_canon_sha256":"5d9989ac7af66daabae9ef68155964cc7249abc615eca14773d2d64efe49343e"},"schema_version":"1.0","source":{"id":"2606.19496","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.19496","created_at":"2026-06-19T16:12:27Z"},{"alias_kind":"arxiv_version","alias_value":"2606.19496v1","created_at":"2026-06-19T16:12:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.19496","created_at":"2026-06-19T16:12:27Z"},{"alias_kind":"pith_short_12","alias_value":"YS43HFSYFHPK","created_at":"2026-06-19T16:12:27Z"},{"alias_kind":"pith_short_16","alias_value":"YS43HFSYFHPKG4LN","created_at":"2026-06-19T16:12:27Z"},{"alias_kind":"pith_short_8","alias_value":"YS43HFSY","created_at":"2026-06-19T16:12:27Z"}],"graph_snapshots":[{"event_id":"sha256:4ac0ab25278ce5035437990c695966b12c9e9729d3f3cb4d58141307e72dfffd","target":"graph","created_at":"2026-06-19T16:12:27Z","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/2606.19496/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Generative models can produce individually plausible samples while deviating substantially from a target set in the distribution of key features. For example, a model pretrained on broad drug-like chemical space may generate molecules whose molecular features differ from those of a therapeutic class of interest, such as known antibiotics. Correcting such distributional miscalibration is challenging: direct finetuning on the target set can overfit and does not control which features are matched. To fill this gap, we introduce kernel Calibrating Generative Models (kCGM). kCGM minimizes a maximum","authors_text":"Brian L. Trippe, Nathaniel L. Diamant","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-17T18:35:16Z","title":"Calibrating Generative Models to Feature Distributions with MMD Finetuning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.19496","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:e1c036a1ac2b17a4defc1df9d3fc7daf0747343b82513ddae24e73aaa11f3c33","target":"record","created_at":"2026-06-19T16:12:27Z","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":"27d42dadd4f9033607b79429281487a7133c03ec956ce33ea6e27598a0ed1173","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-17T18:35:16Z","title_canon_sha256":"5d9989ac7af66daabae9ef68155964cc7249abc615eca14773d2d64efe49343e"},"schema_version":"1.0","source":{"id":"2606.19496","kind":"arxiv","version":1}},"canonical_sha256":"c4b9b3965829dea3716d230b09fad34b658412abb0fddd0bdd4c38fa53c87485","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c4b9b3965829dea3716d230b09fad34b658412abb0fddd0bdd4c38fa53c87485","first_computed_at":"2026-06-19T16:12:27.246797Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-19T16:12:27.246797Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"BehQutSjlOPkKFIiCiIFeAo8x6uPiYtJrZdbMPV/AnugsZCy/ZlOE0pFYHrnNMc6HZHGYyG0OcA2c7qwi7KEBg==","signature_status":"signed_v1","signed_at":"2026-06-19T16:12:27.247142Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.19496","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e1c036a1ac2b17a4defc1df9d3fc7daf0747343b82513ddae24e73aaa11f3c33","sha256:4ac0ab25278ce5035437990c695966b12c9e9729d3f3cb4d58141307e72dfffd"],"state_sha256":"268e328169a5e2353bbdc7945050ff7abaf799bfcc44bab9302c437b375fdb90"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"oq0FHfIJGD/+AW4oRDrFANR4/hMem0fyXGQ7SDUiEjR6XzFQP9PCOxT2H+pxhMQ/6zxEpb0duaXfOZL2DWv3BQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-30T16:18:59.411836Z","bundle_sha256":"8744aac758a4eb4a0774b2df80387cc77b8123f9b232fb3410cabec91d23a7d4"}}