{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2022:BI5GIYZR3AG3LWIOYLW27OGRQO","short_pith_number":"pith:BI5GIYZR","canonical_record":{"source":{"id":"2210.07095","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2022-10-13T15:22:59Z","cross_cats_sorted":[],"title_canon_sha256":"e4b64ef66ee76f675d042c36cad23bfa754704ef8fa3de6a8f18c8d26226a79b","abstract_canon_sha256":"b36b54989642c76360d3272cbd8cad5352164525f8153e70e91006bd042a84da"},"schema_version":"1.0"},"canonical_sha256":"0a3a646331d80db5d90ec2edafb8d183bf4b9fe37171200e1af09bd572f6dbe6","source":{"kind":"arxiv","id":"2210.07095","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2210.07095","created_at":"2026-07-05T05:40:32Z"},{"alias_kind":"arxiv_version","alias_value":"2210.07095v2","created_at":"2026-07-05T05:40:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2210.07095","created_at":"2026-07-05T05:40:32Z"},{"alias_kind":"pith_short_12","alias_value":"BI5GIYZR3AG3","created_at":"2026-07-05T05:40:32Z"},{"alias_kind":"pith_short_16","alias_value":"BI5GIYZR3AG3LWIO","created_at":"2026-07-05T05:40:32Z"},{"alias_kind":"pith_short_8","alias_value":"BI5GIYZR","created_at":"2026-07-05T05:40:32Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2022:BI5GIYZR3AG3LWIOYLW27OGRQO","target":"record","payload":{"canonical_record":{"source":{"id":"2210.07095","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2022-10-13T15:22:59Z","cross_cats_sorted":[],"title_canon_sha256":"e4b64ef66ee76f675d042c36cad23bfa754704ef8fa3de6a8f18c8d26226a79b","abstract_canon_sha256":"b36b54989642c76360d3272cbd8cad5352164525f8153e70e91006bd042a84da"},"schema_version":"1.0"},"canonical_sha256":"0a3a646331d80db5d90ec2edafb8d183bf4b9fe37171200e1af09bd572f6dbe6","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:40:32.523586Z","signature_b64":"A4n6mLpFoDz8u6a1PjhQ5zV94F4cumdioQg5r08j+qUE7GZfr2olHkEsbti5H7RQazYunnFdzH1olNU/dQOMBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0a3a646331d80db5d90ec2edafb8d183bf4b9fe37171200e1af09bd572f6dbe6","last_reissued_at":"2026-07-05T05:40:32.523141Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:40:32.523141Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2210.07095","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":"2026-07-05T05:40:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0agn4mmFDKm2qXaM/S6GPnNK5/0FgjlDmCVGuXG9V5eHAOLbonIo/mVYaWIamYw/gXvfwQp8G/SiEh6Cu+F9CA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T06:26:47.192559Z"},"content_sha256":"4a68686707d19b32528c1e96c19c9a373d33462054f6300c6b8df233d3d41787","schema_version":"1.0","event_id":"sha256:4a68686707d19b32528c1e96c19c9a373d33462054f6300c6b8df233d3d41787"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2022:BI5GIYZR3AG3LWIOYLW27OGRQO","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Incorporating Context into Subword Vocabularies","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Shaked Yehezkel, Yuval Pinter","submitted_at":"2022-10-13T15:22:59Z","abstract_excerpt":"Most current popular subword tokenizers are trained based on word frequency statistics over a corpus, without considering information about co-occurrence or context. Nevertheless, the resulting vocabularies are used in language models' highly contextualized settings. We present SaGe, a tokenizer that tailors subwords for their downstream use by baking in the contextualized signal at the vocabulary creation phase. We show that SaGe does a better job than current widespread tokenizers in keeping token contexts cohesive, while not incurring a large price in terms of encoding efficiency or domain "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2210.07095","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/2210.07095/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":"2026-07-05T05:40:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"2HCMIvy99KWE672MshQ4EDY23Fe2bmXsgWSAlykBTLN7MwxE/y/hNGPk2ZZe3fzmwVpLS8Vs5QNc7GFntRATDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T06:26:47.193058Z"},"content_sha256":"fe89f06e84809c202095d8e18523090e69322b77adeef590d88cd8ecb69b6037","schema_version":"1.0","event_id":"sha256:fe89f06e84809c202095d8e18523090e69322b77adeef590d88cd8ecb69b6037"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BI5GIYZR3AG3LWIOYLW27OGRQO/bundle.json","state_url":"https://pith.science/pith/BI5GIYZR3AG3LWIOYLW27OGRQO/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BI5GIYZR3AG3LWIOYLW27OGRQO/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":"2026-08-18T06:26:47Z","links":{"resolver":"https://pith.science/pith/BI5GIYZR3AG3LWIOYLW27OGRQO","bundle":"https://pith.science/pith/BI5GIYZR3AG3LWIOYLW27OGRQO/bundle.json","state":"https://pith.science/pith/BI5GIYZR3AG3LWIOYLW27OGRQO/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BI5GIYZR3AG3LWIOYLW27OGRQO/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:BI5GIYZR3AG3LWIOYLW27OGRQO","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":"b36b54989642c76360d3272cbd8cad5352164525f8153e70e91006bd042a84da","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2022-10-13T15:22:59Z","title_canon_sha256":"e4b64ef66ee76f675d042c36cad23bfa754704ef8fa3de6a8f18c8d26226a79b"},"schema_version":"1.0","source":{"id":"2210.07095","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2210.07095","created_at":"2026-07-05T05:40:32Z"},{"alias_kind":"arxiv_version","alias_value":"2210.07095v2","created_at":"2026-07-05T05:40:32Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2210.07095","created_at":"2026-07-05T05:40:32Z"},{"alias_kind":"pith_short_12","alias_value":"BI5GIYZR3AG3","created_at":"2026-07-05T05:40:32Z"},{"alias_kind":"pith_short_16","alias_value":"BI5GIYZR3AG3LWIO","created_at":"2026-07-05T05:40:32Z"},{"alias_kind":"pith_short_8","alias_value":"BI5GIYZR","created_at":"2026-07-05T05:40:32Z"}],"graph_snapshots":[{"event_id":"sha256:fe89f06e84809c202095d8e18523090e69322b77adeef590d88cd8ecb69b6037","target":"graph","created_at":"2026-07-05T05:40:32Z","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/2210.07095/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Most current popular subword tokenizers are trained based on word frequency statistics over a corpus, without considering information about co-occurrence or context. Nevertheless, the resulting vocabularies are used in language models' highly contextualized settings. We present SaGe, a tokenizer that tailors subwords for their downstream use by baking in the contextualized signal at the vocabulary creation phase. We show that SaGe does a better job than current widespread tokenizers in keeping token contexts cohesive, while not incurring a large price in terms of encoding efficiency or domain ","authors_text":"Shaked Yehezkel, Yuval Pinter","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2022-10-13T15:22:59Z","title":"Incorporating Context into Subword Vocabularies"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2210.07095","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":"sha256:4a68686707d19b32528c1e96c19c9a373d33462054f6300c6b8df233d3d41787","target":"record","created_at":"2026-07-05T05:40:32Z","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":"b36b54989642c76360d3272cbd8cad5352164525f8153e70e91006bd042a84da","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2022-10-13T15:22:59Z","title_canon_sha256":"e4b64ef66ee76f675d042c36cad23bfa754704ef8fa3de6a8f18c8d26226a79b"},"schema_version":"1.0","source":{"id":"2210.07095","kind":"arxiv","version":2}},"canonical_sha256":"0a3a646331d80db5d90ec2edafb8d183bf4b9fe37171200e1af09bd572f6dbe6","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0a3a646331d80db5d90ec2edafb8d183bf4b9fe37171200e1af09bd572f6dbe6","first_computed_at":"2026-07-05T05:40:32.523141Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:40:32.523141Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"A4n6mLpFoDz8u6a1PjhQ5zV94F4cumdioQg5r08j+qUE7GZfr2olHkEsbti5H7RQazYunnFdzH1olNU/dQOMBw==","signature_status":"signed_v1","signed_at":"2026-07-05T05:40:32.523586Z","signed_message":"canonical_sha256_bytes"},"source_id":"2210.07095","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4a68686707d19b32528c1e96c19c9a373d33462054f6300c6b8df233d3d41787","sha256:fe89f06e84809c202095d8e18523090e69322b77adeef590d88cd8ecb69b6037"],"state_sha256":"622aa8fa4e0d1095ff3e13253f3584016008402912807317f9e9fdfbc00cdfe5"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Uw7PkXNvAuPBcca1Z6QZ7yauP5KIvPmbVndkc2m6+INP+ZznAqJ78E2Aqtb16uWQixbqT49BxU5ejo2qhBK0Ag==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-18T06:26:47.197887Z","bundle_sha256":"e9b8dced1be01595b3d123bcebe73f6d53dac6bb9dff11608281ff4f99691afc"}}