{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:6IW5A6IDQ44L6DJQEQ2KQPXVJZ","short_pith_number":"pith:6IW5A6ID","canonical_record":{"source":{"id":"2501.05078","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-09T09:00:32Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"a3d37bc082d019ad1b965810f613fa9326121d08c2a512873e169d0f0663bb84","abstract_canon_sha256":"2aecdb6dae0f07566be5af92d9bc6286121b2657a8a5835be0ebed8e4e3435a8"},"schema_version":"1.0"},"canonical_sha256":"f22dd079038738bf0d302434a83ef54e749e4ea2bfaa9b6913707fdcc310c633","source":{"kind":"arxiv","id":"2501.05078","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.05078","created_at":"2026-07-05T09:59:04Z"},{"alias_kind":"arxiv_version","alias_value":"2501.05078v1","created_at":"2026-07-05T09:59:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.05078","created_at":"2026-07-05T09:59:04Z"},{"alias_kind":"pith_short_12","alias_value":"6IW5A6IDQ44L","created_at":"2026-07-05T09:59:04Z"},{"alias_kind":"pith_short_16","alias_value":"6IW5A6IDQ44L6DJQ","created_at":"2026-07-05T09:59:04Z"},{"alias_kind":"pith_short_8","alias_value":"6IW5A6ID","created_at":"2026-07-05T09:59:04Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:6IW5A6IDQ44L6DJQEQ2KQPXVJZ","target":"record","payload":{"canonical_record":{"source":{"id":"2501.05078","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-09T09:00:32Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"a3d37bc082d019ad1b965810f613fa9326121d08c2a512873e169d0f0663bb84","abstract_canon_sha256":"2aecdb6dae0f07566be5af92d9bc6286121b2657a8a5835be0ebed8e4e3435a8"},"schema_version":"1.0"},"canonical_sha256":"f22dd079038738bf0d302434a83ef54e749e4ea2bfaa9b6913707fdcc310c633","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:59:04.040985Z","signature_b64":"XQrsGzQQ/WPPhuq46blP/lw58iHmiMMlxVYLeeVCf8vopL3mD90NhGwMvhtEhJP3mADd3dm3G8CkcWJcxReSCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f22dd079038738bf0d302434a83ef54e749e4ea2bfaa9b6913707fdcc310c633","last_reissued_at":"2026-07-05T09:59:04.040545Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:59:04.040545Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2501.05078","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":"2026-07-05T09:59:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wgfIg1GfEUvbEbrerrNWG464XHDwPCtxykeI3qS7NhDwMpewsaNG3/DqQuvQsAZzJBQdTnC0jSoo3r76cfuiAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-13T00:49:51.498687Z"},"content_sha256":"0952c8cd8683f37b97e9225ed236792903a5f41dc7bfb47945da2c6bb4690e00","schema_version":"1.0","event_id":"sha256:0952c8cd8683f37b97e9225ed236792903a5f41dc7bfb47945da2c6bb4690e00"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:6IW5A6IDQ44L6DJQEQ2KQPXVJZ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Analyzing Memorization in Large Language Models through the Lens of Model Attribution","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Chirag Agarwal, Susmit Agrawal, Tarun Ram Menta","submitted_at":"2025-01-09T09:00:32Z","abstract_excerpt":"Large Language Models (LLMs) are prevalent in modern applications but often memorize training data, leading to privacy breaches and copyright issues. Existing research has mainly focused on posthoc analyses, such as extracting memorized content or developing memorization metrics, without exploring the underlying architectural factors that contribute to memorization. In this work, we investigate memorization from an architectural lens by analyzing how attention modules at different layers impact its memorization and generalization performance. Using attribution techniques, we systematically int"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.05078","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/2501.05078/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":"2026-07-05T09:59:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"V9Zp6v4yEqEKDrCetpupkc738RM++H+qqb27QTKRB9ZKq0JBc7n9+UjnMPOgcDlChUwGrU4RL5rPPWoenvWqAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-13T00:49:51.499621Z"},"content_sha256":"de850d7bb8da5e0c6c1aff9759254685da76478dd0c145899f740e5c906a23a1","schema_version":"1.0","event_id":"sha256:de850d7bb8da5e0c6c1aff9759254685da76478dd0c145899f740e5c906a23a1"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/6IW5A6IDQ44L6DJQEQ2KQPXVJZ/bundle.json","state_url":"https://pith.science/pith/6IW5A6IDQ44L6DJQEQ2KQPXVJZ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/6IW5A6IDQ44L6DJQEQ2KQPXVJZ/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":"2026-08-13T00:49:51Z","links":{"resolver":"https://pith.science/pith/6IW5A6IDQ44L6DJQEQ2KQPXVJZ","bundle":"https://pith.science/pith/6IW5A6IDQ44L6DJQEQ2KQPXVJZ/bundle.json","state":"https://pith.science/pith/6IW5A6IDQ44L6DJQEQ2KQPXVJZ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/6IW5A6IDQ44L6DJQEQ2KQPXVJZ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:6IW5A6IDQ44L6DJQEQ2KQPXVJZ","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":"2aecdb6dae0f07566be5af92d9bc6286121b2657a8a5835be0ebed8e4e3435a8","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-09T09:00:32Z","title_canon_sha256":"a3d37bc082d019ad1b965810f613fa9326121d08c2a512873e169d0f0663bb84"},"schema_version":"1.0","source":{"id":"2501.05078","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.05078","created_at":"2026-07-05T09:59:04Z"},{"alias_kind":"arxiv_version","alias_value":"2501.05078v1","created_at":"2026-07-05T09:59:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.05078","created_at":"2026-07-05T09:59:04Z"},{"alias_kind":"pith_short_12","alias_value":"6IW5A6IDQ44L","created_at":"2026-07-05T09:59:04Z"},{"alias_kind":"pith_short_16","alias_value":"6IW5A6IDQ44L6DJQ","created_at":"2026-07-05T09:59:04Z"},{"alias_kind":"pith_short_8","alias_value":"6IW5A6ID","created_at":"2026-07-05T09:59:04Z"}],"graph_snapshots":[{"event_id":"sha256:de850d7bb8da5e0c6c1aff9759254685da76478dd0c145899f740e5c906a23a1","target":"graph","created_at":"2026-07-05T09:59:04Z","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/2501.05078/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Models (LLMs) are prevalent in modern applications but often memorize training data, leading to privacy breaches and copyright issues. Existing research has mainly focused on posthoc analyses, such as extracting memorized content or developing memorization metrics, without exploring the underlying architectural factors that contribute to memorization. In this work, we investigate memorization from an architectural lens by analyzing how attention modules at different layers impact its memorization and generalization performance. Using attribution techniques, we systematically int","authors_text":"Chirag Agarwal, Susmit Agrawal, Tarun Ram Menta","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-09T09:00:32Z","title":"Analyzing Memorization in Large Language Models through the Lens of Model Attribution"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.05078","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":"sha256:0952c8cd8683f37b97e9225ed236792903a5f41dc7bfb47945da2c6bb4690e00","target":"record","created_at":"2026-07-05T09:59:04Z","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":"2aecdb6dae0f07566be5af92d9bc6286121b2657a8a5835be0ebed8e4e3435a8","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-09T09:00:32Z","title_canon_sha256":"a3d37bc082d019ad1b965810f613fa9326121d08c2a512873e169d0f0663bb84"},"schema_version":"1.0","source":{"id":"2501.05078","kind":"arxiv","version":1}},"canonical_sha256":"f22dd079038738bf0d302434a83ef54e749e4ea2bfaa9b6913707fdcc310c633","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f22dd079038738bf0d302434a83ef54e749e4ea2bfaa9b6913707fdcc310c633","first_computed_at":"2026-07-05T09:59:04.040545Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:59:04.040545Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"XQrsGzQQ/WPPhuq46blP/lw58iHmiMMlxVYLeeVCf8vopL3mD90NhGwMvhtEhJP3mADd3dm3G8CkcWJcxReSCw==","signature_status":"signed_v1","signed_at":"2026-07-05T09:59:04.040985Z","signed_message":"canonical_sha256_bytes"},"source_id":"2501.05078","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0952c8cd8683f37b97e9225ed236792903a5f41dc7bfb47945da2c6bb4690e00","sha256:de850d7bb8da5e0c6c1aff9759254685da76478dd0c145899f740e5c906a23a1"],"state_sha256":"9b7667030844bbafa7e823329abf25045e504c68c8e7e20aa42777c6d68e8958"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"eH0jIbm4LpQ4BlVbLUVDb/9z/GbMOIBCRh6rB0VE0ROjE5xWqNMkjp4g7anD/htw+ClEndP0IqlqtXaALPnGAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-13T00:49:51.508617Z","bundle_sha256":"200d4a4b17eafb3d90d47f567b80680d338cd206abe711985877aa901da96579"}}