{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:NY6O7F6IMKZGTD6FTXQMWU57OO","short_pith_number":"pith:NY6O7F6I","canonical_record":{"source":{"id":"2506.06607","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-07T00:51:27Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"aadab502b8a8c27e062a71c28831e6f915216640a1df98138a796a4fd6394d0e","abstract_canon_sha256":"117135cf7642313eb18a2be271852b717844640cab1d0c590b5a2799a125e469"},"schema_version":"1.0"},"canonical_sha256":"6e3cef97c862b2698fc59de0cb53bf73951ea9f0394ae3c934a0b0feb4d47e70","source":{"kind":"arxiv","id":"2506.06607","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.06607","created_at":"2026-07-05T11:17:52Z"},{"alias_kind":"arxiv_version","alias_value":"2506.06607v1","created_at":"2026-07-05T11:17:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.06607","created_at":"2026-07-05T11:17:52Z"},{"alias_kind":"pith_short_12","alias_value":"NY6O7F6IMKZG","created_at":"2026-07-05T11:17:52Z"},{"alias_kind":"pith_short_16","alias_value":"NY6O7F6IMKZGTD6F","created_at":"2026-07-05T11:17:52Z"},{"alias_kind":"pith_short_8","alias_value":"NY6O7F6I","created_at":"2026-07-05T11:17:52Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:NY6O7F6IMKZGTD6FTXQMWU57OO","target":"record","payload":{"canonical_record":{"source":{"id":"2506.06607","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-07T00:51:27Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"aadab502b8a8c27e062a71c28831e6f915216640a1df98138a796a4fd6394d0e","abstract_canon_sha256":"117135cf7642313eb18a2be271852b717844640cab1d0c590b5a2799a125e469"},"schema_version":"1.0"},"canonical_sha256":"6e3cef97c862b2698fc59de0cb53bf73951ea9f0394ae3c934a0b0feb4d47e70","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:17:52.819127Z","signature_b64":"PVCE7sy4B2Jzk4lXAaKPx//SzTO/gDNehtRUFoareAFaaPcRBGjj+li8WQkczuUVgfzy6G9xQvm1xhc4nKDRDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6e3cef97c862b2698fc59de0cb53bf73951ea9f0394ae3c934a0b0feb4d47e70","last_reissued_at":"2026-07-05T11:17:52.818652Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:17:52.818652Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.06607","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":"2026-07-05T11:17:52Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"an0SIEw02LCAFXguHpJ6bQgbQ35crPUMCtpnicjHj+g1i+pJ66zLM2v2Hk6C/cA/BCw13tskic2PSS52KwSGAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T21:05:46.946354Z"},"content_sha256":"6c44c72234911448885ae444d23c10424f2f3305607d73ad862d74731be6c194","schema_version":"1.0","event_id":"sha256:6c44c72234911448885ae444d23c10424f2f3305607d73ad862d74731be6c194"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:NY6O7F6IMKZGTD6FTXQMWU57OO","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Training-Free Tokenizer Transplantation via Orthogonal Matching Pursuit","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Charles Goddard, Fernando Fernandes Neto","submitted_at":"2025-06-07T00:51:27Z","abstract_excerpt":"We present a training-free method to transplant tokenizers in pretrained large language models (LLMs) by reconstructing unseen token embeddings via Orthogonal Matching Pursuit (OMP). Specifically, we approximate each out-of-vocabulary token as a sparse linear combination of shared tokens, in two phases: first, compute each new token's representation in the donor embedding space with a small dictionary of shared anchor tokens, then transfer these same sparse coefficients back into the base model's embedding space.\n  On two challenging cross-tokenizer tasks--Llama$\\to$Mistral NeMo (12B) and Qwen"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.06607","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/2506.06607/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":"2026-07-05T11:17:52Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"eytckY26XTqpVNt4321utyM2Cm1g9/izXeaonfnctS4L7/FaTjC6PJvfiYkaYW8ZbWMndV81zx6yHOQG5l+2BQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T21:05:46.947300Z"},"content_sha256":"4c844e84d4201401f5e4371266fc2c39728596d0b5ea8b561770e38af5f5c52f","schema_version":"1.0","event_id":"sha256:4c844e84d4201401f5e4371266fc2c39728596d0b5ea8b561770e38af5f5c52f"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/NY6O7F6IMKZGTD6FTXQMWU57OO/bundle.json","state_url":"https://pith.science/pith/NY6O7F6IMKZGTD6FTXQMWU57OO/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/NY6O7F6IMKZGTD6FTXQMWU57OO/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":"2026-08-09T21:05:46Z","links":{"resolver":"https://pith.science/pith/NY6O7F6IMKZGTD6FTXQMWU57OO","bundle":"https://pith.science/pith/NY6O7F6IMKZGTD6FTXQMWU57OO/bundle.json","state":"https://pith.science/pith/NY6O7F6IMKZGTD6FTXQMWU57OO/state.json","well_known_bundle":"https://pith.science/.well-known/pith/NY6O7F6IMKZGTD6FTXQMWU57OO/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:NY6O7F6IMKZGTD6FTXQMWU57OO","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":"117135cf7642313eb18a2be271852b717844640cab1d0c590b5a2799a125e469","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-07T00:51:27Z","title_canon_sha256":"aadab502b8a8c27e062a71c28831e6f915216640a1df98138a796a4fd6394d0e"},"schema_version":"1.0","source":{"id":"2506.06607","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.06607","created_at":"2026-07-05T11:17:52Z"},{"alias_kind":"arxiv_version","alias_value":"2506.06607v1","created_at":"2026-07-05T11:17:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.06607","created_at":"2026-07-05T11:17:52Z"},{"alias_kind":"pith_short_12","alias_value":"NY6O7F6IMKZG","created_at":"2026-07-05T11:17:52Z"},{"alias_kind":"pith_short_16","alias_value":"NY6O7F6IMKZGTD6F","created_at":"2026-07-05T11:17:52Z"},{"alias_kind":"pith_short_8","alias_value":"NY6O7F6I","created_at":"2026-07-05T11:17:52Z"}],"graph_snapshots":[{"event_id":"sha256:4c844e84d4201401f5e4371266fc2c39728596d0b5ea8b561770e38af5f5c52f","target":"graph","created_at":"2026-07-05T11:17:52Z","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/2506.06607/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We present a training-free method to transplant tokenizers in pretrained large language models (LLMs) by reconstructing unseen token embeddings via Orthogonal Matching Pursuit (OMP). Specifically, we approximate each out-of-vocabulary token as a sparse linear combination of shared tokens, in two phases: first, compute each new token's representation in the donor embedding space with a small dictionary of shared anchor tokens, then transfer these same sparse coefficients back into the base model's embedding space.\n  On two challenging cross-tokenizer tasks--Llama$\\to$Mistral NeMo (12B) and Qwen","authors_text":"Charles Goddard, Fernando Fernandes Neto","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-07T00:51:27Z","title":"Training-Free Tokenizer Transplantation via Orthogonal Matching Pursuit"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.06607","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":"sha256:6c44c72234911448885ae444d23c10424f2f3305607d73ad862d74731be6c194","target":"record","created_at":"2026-07-05T11:17:52Z","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":"117135cf7642313eb18a2be271852b717844640cab1d0c590b5a2799a125e469","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-07T00:51:27Z","title_canon_sha256":"aadab502b8a8c27e062a71c28831e6f915216640a1df98138a796a4fd6394d0e"},"schema_version":"1.0","source":{"id":"2506.06607","kind":"arxiv","version":1}},"canonical_sha256":"6e3cef97c862b2698fc59de0cb53bf73951ea9f0394ae3c934a0b0feb4d47e70","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6e3cef97c862b2698fc59de0cb53bf73951ea9f0394ae3c934a0b0feb4d47e70","first_computed_at":"2026-07-05T11:17:52.818652Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:17:52.818652Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"PVCE7sy4B2Jzk4lXAaKPx//SzTO/gDNehtRUFoareAFaaPcRBGjj+li8WQkczuUVgfzy6G9xQvm1xhc4nKDRDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:17:52.819127Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.06607","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6c44c72234911448885ae444d23c10424f2f3305607d73ad862d74731be6c194","sha256:4c844e84d4201401f5e4371266fc2c39728596d0b5ea8b561770e38af5f5c52f"],"state_sha256":"f686a46a6ce5fa4ba8adc1e6dddb4b67ceb0093e1299fe0b8bbf487b2684d2fb"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"tbWDvIgl7oy9iel9JWoop9YFwDB3WDs5zyDqQjaKmv5eXn0G23TMKYjQQcwQdm3vO/TOXjrdoV2Y98IRbD3LAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T21:05:46.952353Z","bundle_sha256":"0311b3866ee64dd6aa42fc90fe57ccd5de396890a7c2a9869ab1ae7635a7b908"}}