{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:QO4JZVLI6ZN3KFG33TR3RUKSVY","short_pith_number":"pith:QO4JZVLI","canonical_record":{"source":{"id":"2407.10114","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-14T08:07:50Z","cross_cats_sorted":[],"title_canon_sha256":"8666da57eee6148370a77deaf0a22cbdedce8928130153c81139243decd005b9","abstract_canon_sha256":"adf398e7299f9a25e5b0b0e1c1eeb8e131db194b5446f7e1f56aae5c9c87d5d5"},"schema_version":"1.0"},"canonical_sha256":"83b89cd568f65bb514dbdce3b8d152ae28108c08d5756874ebc6190e6e868687","source":{"kind":"arxiv","id":"2407.10114","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2407.10114","created_at":"2026-07-05T08:46:27Z"},{"alias_kind":"arxiv_version","alias_value":"2407.10114v2","created_at":"2026-07-05T08:46:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.10114","created_at":"2026-07-05T08:46:27Z"},{"alias_kind":"pith_short_12","alias_value":"QO4JZVLI6ZN3","created_at":"2026-07-05T08:46:27Z"},{"alias_kind":"pith_short_16","alias_value":"QO4JZVLI6ZN3KFG3","created_at":"2026-07-05T08:46:27Z"},{"alias_kind":"pith_short_8","alias_value":"QO4JZVLI","created_at":"2026-07-05T08:46:27Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:QO4JZVLI6ZN3KFG33TR3RUKSVY","target":"record","payload":{"canonical_record":{"source":{"id":"2407.10114","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-14T08:07:50Z","cross_cats_sorted":[],"title_canon_sha256":"8666da57eee6148370a77deaf0a22cbdedce8928130153c81139243decd005b9","abstract_canon_sha256":"adf398e7299f9a25e5b0b0e1c1eeb8e131db194b5446f7e1f56aae5c9c87d5d5"},"schema_version":"1.0"},"canonical_sha256":"83b89cd568f65bb514dbdce3b8d152ae28108c08d5756874ebc6190e6e868687","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:46:27.479306Z","signature_b64":"soSAj/AOWHJ830RHb/OsXD8i9YO+X9LmtDalLPG7aAwa4/5kEyHuX88Ql4fQpagbEP54E+rIhbWoY3ESQRtrAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"83b89cd568f65bb514dbdce3b8d152ae28108c08d5756874ebc6190e6e868687","last_reissued_at":"2026-07-05T08:46:27.478868Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:46:27.478868Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2407.10114","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":"2026-07-05T08:46:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"QhUYZjlEYkzfmwBhas/AAHntdGmhmH8S2YNgdj7mpE5/fZBJiYWax7CiJ+jZ3676YJqdPIatc2L5+PG0jC8LAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-21T02:05:33.323517Z"},"content_sha256":"6b763a40e0aeeaa2ab6d717c4dbff8b17c02ed1c06b2e96433c9d7e6742d4628","schema_version":"1.0","event_id":"sha256:6b763a40e0aeeaa2ab6d717c4dbff8b17c02ed1c06b2e96433c9d7e6742d4628"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:QO4JZVLI6ZN3KFG33TR3RUKSVY","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"TokenSHAP: Interpreting Large Language Models with Monte Carlo Shapley Value Estimation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Miriam Horovicz, Roni Goldshmidt","submitted_at":"2024-07-14T08:07:50Z","abstract_excerpt":"As large language models (LLMs) become increasingly prevalent in critical applications, the need for interpretable AI has grown. We introduce TokenSHAP, a novel method for interpreting LLMs by attributing importance to individual tokens or substrings within input prompts. This approach adapts Shapley values from cooperative game theory to natural language processing, offering a rigorous framework for understanding how different parts of an input contribute to a model's response. TokenSHAP leverages Monte Carlo sampling for computational efficiency, providing interpretable, quantitative measure"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.10114","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/2407.10114/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":"2026-07-05T08:46:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rq4pkikf+BBr+gM7kBjK2HOkxkjJH8EvJR7hWO3+5ETh9wVBQ9gkBs2YaLOp3R3cGiFxSge8SGZ0dDInublJDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-21T02:05:33.324062Z"},"content_sha256":"54934c5e55e190a016f98a32eda89e7d47e198094eaacab7a0a3a2e25425dc25","schema_version":"1.0","event_id":"sha256:54934c5e55e190a016f98a32eda89e7d47e198094eaacab7a0a3a2e25425dc25"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/QO4JZVLI6ZN3KFG33TR3RUKSVY/bundle.json","state_url":"https://pith.science/pith/QO4JZVLI6ZN3KFG33TR3RUKSVY/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/QO4JZVLI6ZN3KFG33TR3RUKSVY/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":"2026-08-21T02:05:33Z","links":{"resolver":"https://pith.science/pith/QO4JZVLI6ZN3KFG33TR3RUKSVY","bundle":"https://pith.science/pith/QO4JZVLI6ZN3KFG33TR3RUKSVY/bundle.json","state":"https://pith.science/pith/QO4JZVLI6ZN3KFG33TR3RUKSVY/state.json","well_known_bundle":"https://pith.science/.well-known/pith/QO4JZVLI6ZN3KFG33TR3RUKSVY/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:QO4JZVLI6ZN3KFG33TR3RUKSVY","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":"adf398e7299f9a25e5b0b0e1c1eeb8e131db194b5446f7e1f56aae5c9c87d5d5","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-14T08:07:50Z","title_canon_sha256":"8666da57eee6148370a77deaf0a22cbdedce8928130153c81139243decd005b9"},"schema_version":"1.0","source":{"id":"2407.10114","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2407.10114","created_at":"2026-07-05T08:46:27Z"},{"alias_kind":"arxiv_version","alias_value":"2407.10114v2","created_at":"2026-07-05T08:46:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.10114","created_at":"2026-07-05T08:46:27Z"},{"alias_kind":"pith_short_12","alias_value":"QO4JZVLI6ZN3","created_at":"2026-07-05T08:46:27Z"},{"alias_kind":"pith_short_16","alias_value":"QO4JZVLI6ZN3KFG3","created_at":"2026-07-05T08:46:27Z"},{"alias_kind":"pith_short_8","alias_value":"QO4JZVLI","created_at":"2026-07-05T08:46:27Z"}],"graph_snapshots":[{"event_id":"sha256:54934c5e55e190a016f98a32eda89e7d47e198094eaacab7a0a3a2e25425dc25","target":"graph","created_at":"2026-07-05T08:46:27Z","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/2407.10114/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As large language models (LLMs) become increasingly prevalent in critical applications, the need for interpretable AI has grown. We introduce TokenSHAP, a novel method for interpreting LLMs by attributing importance to individual tokens or substrings within input prompts. This approach adapts Shapley values from cooperative game theory to natural language processing, offering a rigorous framework for understanding how different parts of an input contribute to a model's response. TokenSHAP leverages Monte Carlo sampling for computational efficiency, providing interpretable, quantitative measure","authors_text":"Miriam Horovicz, Roni Goldshmidt","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-14T08:07:50Z","title":"TokenSHAP: Interpreting Large Language Models with Monte Carlo Shapley Value Estimation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.10114","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":"sha256:6b763a40e0aeeaa2ab6d717c4dbff8b17c02ed1c06b2e96433c9d7e6742d4628","target":"record","created_at":"2026-07-05T08:46:27Z","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":"adf398e7299f9a25e5b0b0e1c1eeb8e131db194b5446f7e1f56aae5c9c87d5d5","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-14T08:07:50Z","title_canon_sha256":"8666da57eee6148370a77deaf0a22cbdedce8928130153c81139243decd005b9"},"schema_version":"1.0","source":{"id":"2407.10114","kind":"arxiv","version":2}},"canonical_sha256":"83b89cd568f65bb514dbdce3b8d152ae28108c08d5756874ebc6190e6e868687","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"83b89cd568f65bb514dbdce3b8d152ae28108c08d5756874ebc6190e6e868687","first_computed_at":"2026-07-05T08:46:27.478868Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:46:27.478868Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"soSAj/AOWHJ830RHb/OsXD8i9YO+X9LmtDalLPG7aAwa4/5kEyHuX88Ql4fQpagbEP54E+rIhbWoY3ESQRtrAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T08:46:27.479306Z","signed_message":"canonical_sha256_bytes"},"source_id":"2407.10114","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6b763a40e0aeeaa2ab6d717c4dbff8b17c02ed1c06b2e96433c9d7e6742d4628","sha256:54934c5e55e190a016f98a32eda89e7d47e198094eaacab7a0a3a2e25425dc25"],"state_sha256":"69ef029fd70d3318172cb4996e5e25bdc5eb8861a5f61f6b179f09e92fb6a5c7"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"sxHkJ6wo59jj+2erHoYbO734nglVfJ5xdw+Np/m2JKDz1VpnSK/FnPruAt2b/upY+JKTZygphpFQMQLqXFwJBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-21T02:05:33.328609Z","bundle_sha256":"0bc290326b2c4e0f687b9a204aef0acecf5df2ba47d0671827cfb230d99e1e01"}}