{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:ZZQKGKSLIHRDHK4KLUA4NTGQFL","short_pith_number":"pith:ZZQKGKSL","canonical_record":{"source":{"id":"2406.00509","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-01T17:31:06Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"d59c60ea282cc92b34b0c5dbcca5ff807f8ba8f10eb1247752810662847bd18b","abstract_canon_sha256":"5cebb7f68212d19e0e60a61611b7356f3eea20b17a84e0a487081e7f5a2bb4a0"},"schema_version":"1.0"},"canonical_sha256":"ce60a32a4b41e233ab8a5d01c6ccd02ad8ca4f206b365de1babfbe6dca319d65","source":{"kind":"arxiv","id":"2406.00509","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.00509","created_at":"2026-07-05T08:26:02Z"},{"alias_kind":"arxiv_version","alias_value":"2406.00509v1","created_at":"2026-07-05T08:26:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.00509","created_at":"2026-07-05T08:26:02Z"},{"alias_kind":"pith_short_12","alias_value":"ZZQKGKSLIHRD","created_at":"2026-07-05T08:26:02Z"},{"alias_kind":"pith_short_16","alias_value":"ZZQKGKSLIHRDHK4K","created_at":"2026-07-05T08:26:02Z"},{"alias_kind":"pith_short_8","alias_value":"ZZQKGKSL","created_at":"2026-07-05T08:26:02Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:ZZQKGKSLIHRDHK4KLUA4NTGQFL","target":"record","payload":{"canonical_record":{"source":{"id":"2406.00509","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-01T17:31:06Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"d59c60ea282cc92b34b0c5dbcca5ff807f8ba8f10eb1247752810662847bd18b","abstract_canon_sha256":"5cebb7f68212d19e0e60a61611b7356f3eea20b17a84e0a487081e7f5a2bb4a0"},"schema_version":"1.0"},"canonical_sha256":"ce60a32a4b41e233ab8a5d01c6ccd02ad8ca4f206b365de1babfbe6dca319d65","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:26:02.933435Z","signature_b64":"MSGZuRdjkJBVkAvevoCFJ2GKKMDZKEJMTLm4l782Ocx1rDUgNGOAgUSNg05NKMhT5tym/APskeqxprp3U9ojCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ce60a32a4b41e233ab8a5d01c6ccd02ad8ca4f206b365de1babfbe6dca319d65","last_reissued_at":"2026-07-05T08:26:02.932952Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:26:02.932952Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2406.00509","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":"2026-07-05T08:26:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5Gt54iY0eWNWEWZBtJHxDqSWjMN4MtRRRP4nhAbMpuol+CNvdCMv9jIVdvouA2SI/IePYMWCwlLYunHuAcsgDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-20T16:48:15.530038Z"},"content_sha256":"7bc7a8cbba85b58ab204f16f2ec0237b30c79a42c1636f525adb508f63ef8156","schema_version":"1.0","event_id":"sha256:7bc7a8cbba85b58ab204f16f2ec0237b30c79a42c1636f525adb508f63ef8156"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:ZZQKGKSLIHRDHK4KLUA4NTGQFL","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Empirical influence functions to understand the logic of fine-tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Jordan K. Matelsky, Konrad P. Kording, Lyle Ungar","submitted_at":"2024-06-01T17:31:06Z","abstract_excerpt":"Understanding the process of learning in neural networks is crucial for improving their performance and interpreting their behavior. This can be approximately understood by asking how a model's output is influenced when we fine-tune on a new training sample. There are desiderata for such influences, such as decreasing influence with semantic distance, sparseness, noise invariance, transitive causality, and logical consistency. Here we use the empirical influence measured using fine-tuning to demonstrate how individual training samples affect outputs. We show that these desiderata are violated "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.00509","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/2406.00509/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":"2026-07-05T08:26:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"hb5w7oMtyLfVlPr9JkYyb1uhwu2OszYXz9Fs9u8R48T67rjWd0+umbP8PPyr0iONz360I+ge7ZN/kEokd150AQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-20T16:48:15.530374Z"},"content_sha256":"191b5d1fe29701ef0fe461b983219d21936c2e9f3d62f9317c29fc0324cfbd5e","schema_version":"1.0","event_id":"sha256:191b5d1fe29701ef0fe461b983219d21936c2e9f3d62f9317c29fc0324cfbd5e"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ZZQKGKSLIHRDHK4KLUA4NTGQFL/bundle.json","state_url":"https://pith.science/pith/ZZQKGKSLIHRDHK4KLUA4NTGQFL/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ZZQKGKSLIHRDHK4KLUA4NTGQFL/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":"2026-08-20T16:48:15Z","links":{"resolver":"https://pith.science/pith/ZZQKGKSLIHRDHK4KLUA4NTGQFL","bundle":"https://pith.science/pith/ZZQKGKSLIHRDHK4KLUA4NTGQFL/bundle.json","state":"https://pith.science/pith/ZZQKGKSLIHRDHK4KLUA4NTGQFL/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ZZQKGKSLIHRDHK4KLUA4NTGQFL/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:ZZQKGKSLIHRDHK4KLUA4NTGQFL","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":"5cebb7f68212d19e0e60a61611b7356f3eea20b17a84e0a487081e7f5a2bb4a0","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-01T17:31:06Z","title_canon_sha256":"d59c60ea282cc92b34b0c5dbcca5ff807f8ba8f10eb1247752810662847bd18b"},"schema_version":"1.0","source":{"id":"2406.00509","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.00509","created_at":"2026-07-05T08:26:02Z"},{"alias_kind":"arxiv_version","alias_value":"2406.00509v1","created_at":"2026-07-05T08:26:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.00509","created_at":"2026-07-05T08:26:02Z"},{"alias_kind":"pith_short_12","alias_value":"ZZQKGKSLIHRD","created_at":"2026-07-05T08:26:02Z"},{"alias_kind":"pith_short_16","alias_value":"ZZQKGKSLIHRDHK4K","created_at":"2026-07-05T08:26:02Z"},{"alias_kind":"pith_short_8","alias_value":"ZZQKGKSL","created_at":"2026-07-05T08:26:02Z"}],"graph_snapshots":[{"event_id":"sha256:191b5d1fe29701ef0fe461b983219d21936c2e9f3d62f9317c29fc0324cfbd5e","target":"graph","created_at":"2026-07-05T08:26:02Z","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/2406.00509/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Understanding the process of learning in neural networks is crucial for improving their performance and interpreting their behavior. This can be approximately understood by asking how a model's output is influenced when we fine-tune on a new training sample. There are desiderata for such influences, such as decreasing influence with semantic distance, sparseness, noise invariance, transitive causality, and logical consistency. Here we use the empirical influence measured using fine-tuning to demonstrate how individual training samples affect outputs. We show that these desiderata are violated ","authors_text":"Jordan K. Matelsky, Konrad P. Kording, Lyle Ungar","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-01T17:31:06Z","title":"Empirical influence functions to understand the logic of fine-tuning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.00509","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":"sha256:7bc7a8cbba85b58ab204f16f2ec0237b30c79a42c1636f525adb508f63ef8156","target":"record","created_at":"2026-07-05T08:26:02Z","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":"5cebb7f68212d19e0e60a61611b7356f3eea20b17a84e0a487081e7f5a2bb4a0","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-01T17:31:06Z","title_canon_sha256":"d59c60ea282cc92b34b0c5dbcca5ff807f8ba8f10eb1247752810662847bd18b"},"schema_version":"1.0","source":{"id":"2406.00509","kind":"arxiv","version":1}},"canonical_sha256":"ce60a32a4b41e233ab8a5d01c6ccd02ad8ca4f206b365de1babfbe6dca319d65","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ce60a32a4b41e233ab8a5d01c6ccd02ad8ca4f206b365de1babfbe6dca319d65","first_computed_at":"2026-07-05T08:26:02.932952Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:26:02.932952Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"MSGZuRdjkJBVkAvevoCFJ2GKKMDZKEJMTLm4l782Ocx1rDUgNGOAgUSNg05NKMhT5tym/APskeqxprp3U9ojCg==","signature_status":"signed_v1","signed_at":"2026-07-05T08:26:02.933435Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.00509","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7bc7a8cbba85b58ab204f16f2ec0237b30c79a42c1636f525adb508f63ef8156","sha256:191b5d1fe29701ef0fe461b983219d21936c2e9f3d62f9317c29fc0324cfbd5e"],"state_sha256":"01abc531d4dff5cc159d6eac5cdf86b34e4c514c0f802ceba954425bc7be03c4"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xQFT9I83fSxJLUEEZhbGsyrvcS1dhWStZfemuNx/ZZWIe1sOvcTc55b8+VgFxbMU/4KNobdGgxgUxPEPHn9vCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-20T16:48:15.533996Z","bundle_sha256":"172dd9e6b6841c9e303e85953f3ad575ad0e0e5a9a9b64fc650d05e4e9ea4d75"}}