{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:K5I7V237CXQSIU6W7SSEYWUFMM","short_pith_number":"pith:K5I7V237","canonical_record":{"source":{"id":"2412.14528","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-12-19T04:51:06Z","cross_cats_sorted":[],"title_canon_sha256":"fe3e787c50dffc08566a39f43a5d99ae112e3ec78167017456cb5bdcfc6d9b67","abstract_canon_sha256":"6952fdc5abeb8028eb9e6ecb707be2fcbce49347b501fc14352ab4f2eaaf01cc"},"schema_version":"1.0"},"canonical_sha256":"5751faeb7f15e12453d6fca44c5a85632d6d076b7b938cefa1d6868f37198d3c","source":{"kind":"arxiv","id":"2412.14528","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.14528","created_at":"2026-07-05T10:02:25Z"},{"alias_kind":"arxiv_version","alias_value":"2412.14528v2","created_at":"2026-07-05T10:02:25Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.14528","created_at":"2026-07-05T10:02:25Z"},{"alias_kind":"pith_short_12","alias_value":"K5I7V237CXQS","created_at":"2026-07-05T10:02:25Z"},{"alias_kind":"pith_short_16","alias_value":"K5I7V237CXQSIU6W","created_at":"2026-07-05T10:02:25Z"},{"alias_kind":"pith_short_8","alias_value":"K5I7V237","created_at":"2026-07-05T10:02:25Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:K5I7V237CXQSIU6W7SSEYWUFMM","target":"record","payload":{"canonical_record":{"source":{"id":"2412.14528","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-12-19T04:51:06Z","cross_cats_sorted":[],"title_canon_sha256":"fe3e787c50dffc08566a39f43a5d99ae112e3ec78167017456cb5bdcfc6d9b67","abstract_canon_sha256":"6952fdc5abeb8028eb9e6ecb707be2fcbce49347b501fc14352ab4f2eaaf01cc"},"schema_version":"1.0"},"canonical_sha256":"5751faeb7f15e12453d6fca44c5a85632d6d076b7b938cefa1d6868f37198d3c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:02:25.783609Z","signature_b64":"UQ4tQZljj3a/HyzNBEb0aPgSKMOsm5UIcsxKb5okl/y4A9xAN87+fTmQN0xvd5laush+ewvq24+P4dxa6K/5AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5751faeb7f15e12453d6fca44c5a85632d6d076b7b938cefa1d6868f37198d3c","last_reissued_at":"2026-07-05T10:02:25.783164Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:02:25.783164Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2412.14528","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":"2026-07-05T10:02:25Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"XGUXcdXKwrxEjestUYm4XjJSJup8rqrmabi9vmUCn6H8Xul9am6ib7wU7bARYV3YeiJUYZ0JTyvqJJcNH/loAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T03:23:25.522598Z"},"content_sha256":"e15b284b30c66c8e6074d53170aa628a52a80585b8b509c14cfd2703818b6cb9","schema_version":"1.0","event_id":"sha256:e15b284b30c66c8e6074d53170aa628a52a80585b8b509c14cfd2703818b6cb9"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:K5I7V237CXQSIU6W7SSEYWUFMM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Multi-Level Optimal Transport for Universal Cross-Tokenizer Knowledge Distillation on Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Houqiang Li, Liang Xie, Mo Zhu, Wengang Zhou, Xiao Cui, Yulei Qin","submitted_at":"2024-12-19T04:51:06Z","abstract_excerpt":"Knowledge distillation (KD) has become a prevalent technique for compressing large language models (LLMs). Existing KD methods are constrained by the need for identical tokenizers (i.e., vocabularies) between teacher and student models, limiting their versatility in handling LLMs of different architecture families. In this paper, we introduce the Multi-Level Optimal Transport (MultiLevelOT), a novel approach that advances the optimal transport for universal cross-tokenizer knowledge distillation. Our method aligns the logit distributions of the teacher and the student at both token and sequenc"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.14528","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/2412.14528/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":"2026-07-05T10:02:25Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mbCb4hCCdYhJAukXkE7ekmFmWIpgMYrgWv8ey+p30T3KmjM11gzpeARNM7IgHwSb3v+IGU153RpU27DETpEVAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T03:23:25.522984Z"},"content_sha256":"1f808bc9b5fe69440d1aa809421af1f8d20481f013e00a4c2e524d0855004103","schema_version":"1.0","event_id":"sha256:1f808bc9b5fe69440d1aa809421af1f8d20481f013e00a4c2e524d0855004103"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/K5I7V237CXQSIU6W7SSEYWUFMM/bundle.json","state_url":"https://pith.science/pith/K5I7V237CXQSIU6W7SSEYWUFMM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/K5I7V237CXQSIU6W7SSEYWUFMM/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":"2026-08-14T03:23:25Z","links":{"resolver":"https://pith.science/pith/K5I7V237CXQSIU6W7SSEYWUFMM","bundle":"https://pith.science/pith/K5I7V237CXQSIU6W7SSEYWUFMM/bundle.json","state":"https://pith.science/pith/K5I7V237CXQSIU6W7SSEYWUFMM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/K5I7V237CXQSIU6W7SSEYWUFMM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:K5I7V237CXQSIU6W7SSEYWUFMM","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":"6952fdc5abeb8028eb9e6ecb707be2fcbce49347b501fc14352ab4f2eaaf01cc","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-12-19T04:51:06Z","title_canon_sha256":"fe3e787c50dffc08566a39f43a5d99ae112e3ec78167017456cb5bdcfc6d9b67"},"schema_version":"1.0","source":{"id":"2412.14528","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.14528","created_at":"2026-07-05T10:02:25Z"},{"alias_kind":"arxiv_version","alias_value":"2412.14528v2","created_at":"2026-07-05T10:02:25Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.14528","created_at":"2026-07-05T10:02:25Z"},{"alias_kind":"pith_short_12","alias_value":"K5I7V237CXQS","created_at":"2026-07-05T10:02:25Z"},{"alias_kind":"pith_short_16","alias_value":"K5I7V237CXQSIU6W","created_at":"2026-07-05T10:02:25Z"},{"alias_kind":"pith_short_8","alias_value":"K5I7V237","created_at":"2026-07-05T10:02:25Z"}],"graph_snapshots":[{"event_id":"sha256:1f808bc9b5fe69440d1aa809421af1f8d20481f013e00a4c2e524d0855004103","target":"graph","created_at":"2026-07-05T10:02:25Z","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/2412.14528/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Knowledge distillation (KD) has become a prevalent technique for compressing large language models (LLMs). Existing KD methods are constrained by the need for identical tokenizers (i.e., vocabularies) between teacher and student models, limiting their versatility in handling LLMs of different architecture families. In this paper, we introduce the Multi-Level Optimal Transport (MultiLevelOT), a novel approach that advances the optimal transport for universal cross-tokenizer knowledge distillation. Our method aligns the logit distributions of the teacher and the student at both token and sequenc","authors_text":"Houqiang Li, Liang Xie, Mo Zhu, Wengang Zhou, Xiao Cui, Yulei Qin","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-12-19T04:51:06Z","title":"Multi-Level Optimal Transport for Universal Cross-Tokenizer Knowledge Distillation on Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.14528","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:e15b284b30c66c8e6074d53170aa628a52a80585b8b509c14cfd2703818b6cb9","target":"record","created_at":"2026-07-05T10:02:25Z","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":"6952fdc5abeb8028eb9e6ecb707be2fcbce49347b501fc14352ab4f2eaaf01cc","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-12-19T04:51:06Z","title_canon_sha256":"fe3e787c50dffc08566a39f43a5d99ae112e3ec78167017456cb5bdcfc6d9b67"},"schema_version":"1.0","source":{"id":"2412.14528","kind":"arxiv","version":2}},"canonical_sha256":"5751faeb7f15e12453d6fca44c5a85632d6d076b7b938cefa1d6868f37198d3c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5751faeb7f15e12453d6fca44c5a85632d6d076b7b938cefa1d6868f37198d3c","first_computed_at":"2026-07-05T10:02:25.783164Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:02:25.783164Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"UQ4tQZljj3a/HyzNBEb0aPgSKMOsm5UIcsxKb5okl/y4A9xAN87+fTmQN0xvd5laush+ewvq24+P4dxa6K/5AQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:02:25.783609Z","signed_message":"canonical_sha256_bytes"},"source_id":"2412.14528","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e15b284b30c66c8e6074d53170aa628a52a80585b8b509c14cfd2703818b6cb9","sha256:1f808bc9b5fe69440d1aa809421af1f8d20481f013e00a4c2e524d0855004103"],"state_sha256":"bd9a7be61c124fc0ba56ecdd7c03c031421627b1fe2046c4d0c76ba40f95740c"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"BKAzwciSzHMPOgiVexXkSnyHvtZcSRFIfLDo6NkFmz+99LrlUS5OBKVRrtzptGYb5uguAKpafaFHpl9sZFQ6BA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-14T03:23:25.526038Z","bundle_sha256":"b23ecd9e439f4434e72c36468c712d609f5ba701320d4b4880ad634989d5d225"}}