{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:BFPARLCJRVUUZSK3GCATOBWAU4","short_pith_number":"pith:BFPARLCJ","canonical_record":{"source":{"id":"2412.03886","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-05T05:39:03Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b43356258a4d3f03323fac34e1387423853c743e5b2acd96b70ac6fa34c4a145","abstract_canon_sha256":"b669ab767ab2eca3b22f2d7496131d8ef87038d7f300f26fe67dc2b28dab733f"},"schema_version":"1.0"},"canonical_sha256":"095e08ac498d694cc95b30813706c0a73f2e089c7a0eafe65b283ac3459d1961","source":{"kind":"arxiv","id":"2412.03886","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.03886","created_at":"2026-07-05T09:44:55Z"},{"alias_kind":"arxiv_version","alias_value":"2412.03886v1","created_at":"2026-07-05T09:44:55Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.03886","created_at":"2026-07-05T09:44:55Z"},{"alias_kind":"pith_short_12","alias_value":"BFPARLCJRVUU","created_at":"2026-07-05T09:44:55Z"},{"alias_kind":"pith_short_16","alias_value":"BFPARLCJRVUUZSK3","created_at":"2026-07-05T09:44:55Z"},{"alias_kind":"pith_short_8","alias_value":"BFPARLCJ","created_at":"2026-07-05T09:44:55Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:BFPARLCJRVUUZSK3GCATOBWAU4","target":"record","payload":{"canonical_record":{"source":{"id":"2412.03886","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-05T05:39:03Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b43356258a4d3f03323fac34e1387423853c743e5b2acd96b70ac6fa34c4a145","abstract_canon_sha256":"b669ab767ab2eca3b22f2d7496131d8ef87038d7f300f26fe67dc2b28dab733f"},"schema_version":"1.0"},"canonical_sha256":"095e08ac498d694cc95b30813706c0a73f2e089c7a0eafe65b283ac3459d1961","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:44:55.858196Z","signature_b64":"HSTg8MPf5okRAI0JJ7PFtdJ3eK9j8Ze9C2+Z5YORwQ6QqfBnrniAcjzWKykUQs5CWrKuHbxOHdIrPzIIRIpkCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"095e08ac498d694cc95b30813706c0a73f2e089c7a0eafe65b283ac3459d1961","last_reissued_at":"2026-07-05T09:44:55.857756Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:44:55.857756Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2412.03886","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":"2026-07-05T09:44:55Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"WtW9lvD/0OvisFRXMKK4AbvvgJTxZ6WcnJBumuDrenlVeN7bsTfdoTpZGV1qBM3bNKC7n00wijqF8J1arV3QDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-19T17:35:01.020966Z"},"content_sha256":"bd5d6ade47311fe6cc3a88de0bf1b1254cba61dd3be2d2c0b4a37191f28ccc22","schema_version":"1.0","event_id":"sha256:bd5d6ade47311fe6cc3a88de0bf1b1254cba61dd3be2d2c0b4a37191f28ccc22"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:BFPARLCJRVUUZSK3GCATOBWAU4","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Uniform Discretized Integrated Gradients: An effective attribution based method for explaining large language models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Ayan Kundu, Swarnava Sinha Roy","submitted_at":"2024-12-05T05:39:03Z","abstract_excerpt":"Integrated Gradients is a well-known technique for explaining deep learning models. It calculates feature importance scores by employing a gradient based approach computing gradients of the model output with respect to input features and accumulating them along a linear path. While this works well for continuous features spaces, it may not be the most optimal way to deal with discrete spaces like word embeddings. For interpreting LLMs (Large Language Models), there exists a need for a non-linear path where intermediate points, whose gradients are to be computed, lie close to actual words in th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.03886","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/2412.03886/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":"2026-07-05T09:44:55Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ENJWiLmFl7iHivuSQO7o7wbYDAPyPqeO3Zjk3eey2iF4IO8VKSxX2ZRV2eQYzqBLCddV66HkXshw/HAGNLzgBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-19T17:35:01.022001Z"},"content_sha256":"6801d0b452d7fe29d61b26f3d668e312bc714649d5d015a3908cd30e61c8d429","schema_version":"1.0","event_id":"sha256:6801d0b452d7fe29d61b26f3d668e312bc714649d5d015a3908cd30e61c8d429"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BFPARLCJRVUUZSK3GCATOBWAU4/bundle.json","state_url":"https://pith.science/pith/BFPARLCJRVUUZSK3GCATOBWAU4/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BFPARLCJRVUUZSK3GCATOBWAU4/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":"2026-08-19T17:35:01Z","links":{"resolver":"https://pith.science/pith/BFPARLCJRVUUZSK3GCATOBWAU4","bundle":"https://pith.science/pith/BFPARLCJRVUUZSK3GCATOBWAU4/bundle.json","state":"https://pith.science/pith/BFPARLCJRVUUZSK3GCATOBWAU4/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BFPARLCJRVUUZSK3GCATOBWAU4/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:BFPARLCJRVUUZSK3GCATOBWAU4","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":"b669ab767ab2eca3b22f2d7496131d8ef87038d7f300f26fe67dc2b28dab733f","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-05T05:39:03Z","title_canon_sha256":"b43356258a4d3f03323fac34e1387423853c743e5b2acd96b70ac6fa34c4a145"},"schema_version":"1.0","source":{"id":"2412.03886","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.03886","created_at":"2026-07-05T09:44:55Z"},{"alias_kind":"arxiv_version","alias_value":"2412.03886v1","created_at":"2026-07-05T09:44:55Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.03886","created_at":"2026-07-05T09:44:55Z"},{"alias_kind":"pith_short_12","alias_value":"BFPARLCJRVUU","created_at":"2026-07-05T09:44:55Z"},{"alias_kind":"pith_short_16","alias_value":"BFPARLCJRVUUZSK3","created_at":"2026-07-05T09:44:55Z"},{"alias_kind":"pith_short_8","alias_value":"BFPARLCJ","created_at":"2026-07-05T09:44:55Z"}],"graph_snapshots":[{"event_id":"sha256:6801d0b452d7fe29d61b26f3d668e312bc714649d5d015a3908cd30e61c8d429","target":"graph","created_at":"2026-07-05T09:44:55Z","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/2412.03886/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Integrated Gradients is a well-known technique for explaining deep learning models. It calculates feature importance scores by employing a gradient based approach computing gradients of the model output with respect to input features and accumulating them along a linear path. While this works well for continuous features spaces, it may not be the most optimal way to deal with discrete spaces like word embeddings. For interpreting LLMs (Large Language Models), there exists a need for a non-linear path where intermediate points, whose gradients are to be computed, lie close to actual words in th","authors_text":"Ayan Kundu, Swarnava Sinha Roy","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-05T05:39:03Z","title":"Uniform Discretized Integrated Gradients: An effective attribution based method for explaining large language models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.03886","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:bd5d6ade47311fe6cc3a88de0bf1b1254cba61dd3be2d2c0b4a37191f28ccc22","target":"record","created_at":"2026-07-05T09:44:55Z","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":"b669ab767ab2eca3b22f2d7496131d8ef87038d7f300f26fe67dc2b28dab733f","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-05T05:39:03Z","title_canon_sha256":"b43356258a4d3f03323fac34e1387423853c743e5b2acd96b70ac6fa34c4a145"},"schema_version":"1.0","source":{"id":"2412.03886","kind":"arxiv","version":1}},"canonical_sha256":"095e08ac498d694cc95b30813706c0a73f2e089c7a0eafe65b283ac3459d1961","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"095e08ac498d694cc95b30813706c0a73f2e089c7a0eafe65b283ac3459d1961","first_computed_at":"2026-07-05T09:44:55.857756Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:44:55.857756Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"HSTg8MPf5okRAI0JJ7PFtdJ3eK9j8Ze9C2+Z5YORwQ6QqfBnrniAcjzWKykUQs5CWrKuHbxOHdIrPzIIRIpkCA==","signature_status":"signed_v1","signed_at":"2026-07-05T09:44:55.858196Z","signed_message":"canonical_sha256_bytes"},"source_id":"2412.03886","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:bd5d6ade47311fe6cc3a88de0bf1b1254cba61dd3be2d2c0b4a37191f28ccc22","sha256:6801d0b452d7fe29d61b26f3d668e312bc714649d5d015a3908cd30e61c8d429"],"state_sha256":"59f56edca49b7425d7b5363d314a0ec59980e611769a29ee77866165dfc0a493"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"uTWw+M/UBpW3u5oi239RtFIVQmCHBgqvVrShePJe0KgUC2+tM0khY8eOJDBhGztOjd0kePSkN/8Y0CZtrIydAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-19T17:35:01.031925Z","bundle_sha256":"0a0a87fc1f41e1b6c6f1c3ac8958d300bb3c5b86448749dfa2456e616da05751"}}