{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:4OTQOL2JZ5JT7B2Y5ZOLTOYGI7","short_pith_number":"pith:4OTQOL2J","canonical_record":{"source":{"id":"2406.14561","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-20T17:59:42Z","cross_cats_sorted":[],"title_canon_sha256":"2b846bc106320598cae3de8040e3a4cadf6531657606be8d80959cc5d19ca9d9","abstract_canon_sha256":"e47bc90d8f3649aeebc2b7fe80471fce13f3e25204c9477cd83125c4c1b5d44c"},"schema_version":"1.0"},"canonical_sha256":"e3a7072f49cf533f8758ee5cb9bb0647cc6d4b051be9bcc869c322647fa14c01","source":{"kind":"arxiv","id":"2406.14561","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.14561","created_at":"2026-07-05T09:19:29Z"},{"alias_kind":"arxiv_version","alias_value":"2406.14561v2","created_at":"2026-07-05T09:19:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.14561","created_at":"2026-07-05T09:19:29Z"},{"alias_kind":"pith_short_12","alias_value":"4OTQOL2JZ5JT","created_at":"2026-07-05T09:19:29Z"},{"alias_kind":"pith_short_16","alias_value":"4OTQOL2JZ5JT7B2Y","created_at":"2026-07-05T09:19:29Z"},{"alias_kind":"pith_short_8","alias_value":"4OTQOL2J","created_at":"2026-07-05T09:19:29Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:4OTQOL2JZ5JT7B2Y5ZOLTOYGI7","target":"record","payload":{"canonical_record":{"source":{"id":"2406.14561","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-20T17:59:42Z","cross_cats_sorted":[],"title_canon_sha256":"2b846bc106320598cae3de8040e3a4cadf6531657606be8d80959cc5d19ca9d9","abstract_canon_sha256":"e47bc90d8f3649aeebc2b7fe80471fce13f3e25204c9477cd83125c4c1b5d44c"},"schema_version":"1.0"},"canonical_sha256":"e3a7072f49cf533f8758ee5cb9bb0647cc6d4b051be9bcc869c322647fa14c01","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:19:29.511151Z","signature_b64":"0HRsN3prUCGlsu8oi/yDClT3aR/ldY78mEWu3uu2FY8v9NTTgIW3DS/ZPNGYPmdsMF4FvYss4WOufEX3YKsVDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e3a7072f49cf533f8758ee5cb9bb0647cc6d4b051be9bcc869c322647fa14c01","last_reissued_at":"2026-07-05T09:19:29.510641Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:19:29.510641Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2406.14561","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":"2026-07-05T09:19:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"2Obf/A13o4WGRgRkeLdBrQ6HOALBL+inRH15mV69BXbEbKJC27jhkASCnx1wMth0IKPLC4NPHFqoth2G9K2ODA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-21T02:45:02.394892Z"},"content_sha256":"47742580053b2c412a1690fbe890d909e539389ee7b415574acd379f168f3362","schema_version":"1.0","event_id":"sha256:47742580053b2c412a1690fbe890d909e539389ee7b415574acd379f168f3362"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:4OTQOL2JZ5JT7B2Y5ZOLTOYGI7","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"How to Compute the Probability of a Word","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Clara Meister, Tiago Pimentel","submitted_at":"2024-06-20T17:59:42Z","abstract_excerpt":"Language models (LMs) estimate a probability distribution over strings in a natural language; these distributions are crucial for computing perplexity and surprisal in linguistics research. While we are usually concerned with measuring these values for words, most LMs operate over subwords. Despite seemingly straightforward, accurately computing probabilities over one unit given probabilities over the other requires care. Indeed, we show here that many recent linguistic studies have been incorrectly computing these values. This paper derives the correct methods for computing word probabilities"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.14561","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/2406.14561/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":"2026-07-05T09:19:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"fP1CsisN02HC6srHbVQqT/lE/krFR/iOcOcSvmqJMA8Nc7+g5SfjSLnYTg0sgT+zFw8XZViXcaPK2/+W7IdgBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-21T02:45:02.395398Z"},"content_sha256":"1f1e059700c7e5027809b48c80be38a0c61d82d09c7fb05bb87a11bb07a24f27","schema_version":"1.0","event_id":"sha256:1f1e059700c7e5027809b48c80be38a0c61d82d09c7fb05bb87a11bb07a24f27"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4OTQOL2JZ5JT7B2Y5ZOLTOYGI7/bundle.json","state_url":"https://pith.science/pith/4OTQOL2JZ5JT7B2Y5ZOLTOYGI7/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4OTQOL2JZ5JT7B2Y5ZOLTOYGI7/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":"2026-08-21T02:45:02Z","links":{"resolver":"https://pith.science/pith/4OTQOL2JZ5JT7B2Y5ZOLTOYGI7","bundle":"https://pith.science/pith/4OTQOL2JZ5JT7B2Y5ZOLTOYGI7/bundle.json","state":"https://pith.science/pith/4OTQOL2JZ5JT7B2Y5ZOLTOYGI7/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4OTQOL2JZ5JT7B2Y5ZOLTOYGI7/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:4OTQOL2JZ5JT7B2Y5ZOLTOYGI7","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":"e47bc90d8f3649aeebc2b7fe80471fce13f3e25204c9477cd83125c4c1b5d44c","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-20T17:59:42Z","title_canon_sha256":"2b846bc106320598cae3de8040e3a4cadf6531657606be8d80959cc5d19ca9d9"},"schema_version":"1.0","source":{"id":"2406.14561","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.14561","created_at":"2026-07-05T09:19:29Z"},{"alias_kind":"arxiv_version","alias_value":"2406.14561v2","created_at":"2026-07-05T09:19:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.14561","created_at":"2026-07-05T09:19:29Z"},{"alias_kind":"pith_short_12","alias_value":"4OTQOL2JZ5JT","created_at":"2026-07-05T09:19:29Z"},{"alias_kind":"pith_short_16","alias_value":"4OTQOL2JZ5JT7B2Y","created_at":"2026-07-05T09:19:29Z"},{"alias_kind":"pith_short_8","alias_value":"4OTQOL2J","created_at":"2026-07-05T09:19:29Z"}],"graph_snapshots":[{"event_id":"sha256:1f1e059700c7e5027809b48c80be38a0c61d82d09c7fb05bb87a11bb07a24f27","target":"graph","created_at":"2026-07-05T09:19:29Z","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/2406.14561/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Language models (LMs) estimate a probability distribution over strings in a natural language; these distributions are crucial for computing perplexity and surprisal in linguistics research. While we are usually concerned with measuring these values for words, most LMs operate over subwords. Despite seemingly straightforward, accurately computing probabilities over one unit given probabilities over the other requires care. Indeed, we show here that many recent linguistic studies have been incorrectly computing these values. This paper derives the correct methods for computing word probabilities","authors_text":"Clara Meister, Tiago Pimentel","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-20T17:59:42Z","title":"How to Compute the Probability of a Word"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.14561","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":"sha256:47742580053b2c412a1690fbe890d909e539389ee7b415574acd379f168f3362","target":"record","created_at":"2026-07-05T09:19:29Z","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":"e47bc90d8f3649aeebc2b7fe80471fce13f3e25204c9477cd83125c4c1b5d44c","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-06-20T17:59:42Z","title_canon_sha256":"2b846bc106320598cae3de8040e3a4cadf6531657606be8d80959cc5d19ca9d9"},"schema_version":"1.0","source":{"id":"2406.14561","kind":"arxiv","version":2}},"canonical_sha256":"e3a7072f49cf533f8758ee5cb9bb0647cc6d4b051be9bcc869c322647fa14c01","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e3a7072f49cf533f8758ee5cb9bb0647cc6d4b051be9bcc869c322647fa14c01","first_computed_at":"2026-07-05T09:19:29.510641Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:19:29.510641Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"0HRsN3prUCGlsu8oi/yDClT3aR/ldY78mEWu3uu2FY8v9NTTgIW3DS/ZPNGYPmdsMF4FvYss4WOufEX3YKsVDA==","signature_status":"signed_v1","signed_at":"2026-07-05T09:19:29.511151Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.14561","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:47742580053b2c412a1690fbe890d909e539389ee7b415574acd379f168f3362","sha256:1f1e059700c7e5027809b48c80be38a0c61d82d09c7fb05bb87a11bb07a24f27"],"state_sha256":"f5c7b11c01ebf34f4a293c927d88362169136151fa168fb640df1220874618c4"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"vZjZJAVKzAryOtkSWOVnwBrLRxc2TLP+UL4ImF6YrKp0GgWGW8aWyApx02k37a4W2oexJer7tinXnT31ImQnBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-21T02:45:02.452987Z","bundle_sha256":"9bd72940254f9d07863d34742a6d79aee6f3b74d2895d6938036d173c16467d9"}}