{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:ILYMW5VHJOXJD6GTEHWV5PIJEA","short_pith_number":"pith:ILYMW5VH","canonical_record":{"source":{"id":"2603.22056","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2026-03-23T14:53:55Z","cross_cats_sorted":[],"title_canon_sha256":"0097e9376adcf1b7fdf3d3742de173b12760ef9d4d6cfcf208aff17f87cc1b00","abstract_canon_sha256":"dad5156d43b04df52db3451cf11332d9868c3f58a0b8e15dee5c7d694e7e24c9"},"schema_version":"1.0"},"canonical_sha256":"42f0cb76a74bae91f8d321ed5ebd0920027dc9c07599799d843a418a69b7d381","source":{"kind":"arxiv","id":"2603.22056","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2603.22056","created_at":"2026-05-20T00:04:29Z"},{"alias_kind":"arxiv_version","alias_value":"2603.22056v2","created_at":"2026-05-20T00:04:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.22056","created_at":"2026-05-20T00:04:29Z"},{"alias_kind":"pith_short_12","alias_value":"ILYMW5VHJOXJ","created_at":"2026-05-20T00:04:29Z"},{"alias_kind":"pith_short_16","alias_value":"ILYMW5VHJOXJD6GT","created_at":"2026-05-20T00:04:29Z"},{"alias_kind":"pith_short_8","alias_value":"ILYMW5VH","created_at":"2026-05-20T00:04:29Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:ILYMW5VHJOXJD6GTEHWV5PIJEA","target":"record","payload":{"canonical_record":{"source":{"id":"2603.22056","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2026-03-23T14:53:55Z","cross_cats_sorted":[],"title_canon_sha256":"0097e9376adcf1b7fdf3d3742de173b12760ef9d4d6cfcf208aff17f87cc1b00","abstract_canon_sha256":"dad5156d43b04df52db3451cf11332d9868c3f58a0b8e15dee5c7d694e7e24c9"},"schema_version":"1.0"},"canonical_sha256":"42f0cb76a74bae91f8d321ed5ebd0920027dc9c07599799d843a418a69b7d381","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:04:29.140986Z","signature_b64":"wVgkawp3R+3xmZxqrN9EFZmV99Qgymln/NdrrKAFztTRHL1EVZcvh5EEHUR0a3NTF98jMCWYfaYw9VIBMxPUCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"42f0cb76a74bae91f8d321ed5ebd0920027dc9c07599799d843a418a69b7d381","last_reissued_at":"2026-05-20T00:04:29.139895Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:04:29.139895Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2603.22056","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":"2026-05-20T00:04:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"iGpa7qQlmiAAjuuYDo2o8C6YeVN4hVs/ivEuH/TzOjkFtFoxbXhFWqwitOma5ASG/2BAhB6g+Nw6ZXjEpsKOAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T06:35:26.327844Z"},"content_sha256":"3118186f1d12baf1a20f521f29d1f1d1698936fb130ea5c7317fd5d61a0f2a13","schema_version":"1.0","event_id":"sha256:3118186f1d12baf1a20f521f29d1f1d1698936fb130ea5c7317fd5d61a0f2a13"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:ILYMW5VHJOXJD6GTEHWV5PIJEA","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Dual-Space Knowledge Distillation with Key-Query Matching for Large Language Models with Vocabulary Mismatch","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Cong-Thanh Do, Kate Knill, Stella Eva Tsiapali","submitted_at":"2026-03-23T14:53:55Z","abstract_excerpt":"Large language models (LLMs) achieve state-of-the-art (SOTA) performance across language tasks, but are costly to deploy due to their size and resource demands. Knowledge Distillation (KD) addresses this by training smaller Student models to mimic larger Teacher models, improving efficiency without significant performance loss. Dual-Space Knowledge Distillation with Cross-Model Attention (DSKD-CMA) has emerged as a SOTA method for KD between LLMs with distinct tokenizers, yet its internal workings remain largely opaque. In this work, we systematically analyse the attention mechanism of DSKD-CM"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.22056","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/2603.22056/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":"2026-05-20T00:04:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"WRiufUfHebKN6TxxqclR6VtvEeTTvMbRlcO1x44L6GPLzJN777dGvV7P6+RMPXnrrEhNCryMfrMpviDVDh5vBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T06:35:26.328443Z"},"content_sha256":"6d767527ff35fe8016735807375de816cefc6681d88d8626b320103bc63ce0ee","schema_version":"1.0","event_id":"sha256:6d767527ff35fe8016735807375de816cefc6681d88d8626b320103bc63ce0ee"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ILYMW5VHJOXJD6GTEHWV5PIJEA/bundle.json","state_url":"https://pith.science/pith/ILYMW5VHJOXJD6GTEHWV5PIJEA/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ILYMW5VHJOXJD6GTEHWV5PIJEA/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":"2026-05-26T06:35:26Z","links":{"resolver":"https://pith.science/pith/ILYMW5VHJOXJD6GTEHWV5PIJEA","bundle":"https://pith.science/pith/ILYMW5VHJOXJD6GTEHWV5PIJEA/bundle.json","state":"https://pith.science/pith/ILYMW5VHJOXJD6GTEHWV5PIJEA/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ILYMW5VHJOXJD6GTEHWV5PIJEA/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:ILYMW5VHJOXJD6GTEHWV5PIJEA","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":"dad5156d43b04df52db3451cf11332d9868c3f58a0b8e15dee5c7d694e7e24c9","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2026-03-23T14:53:55Z","title_canon_sha256":"0097e9376adcf1b7fdf3d3742de173b12760ef9d4d6cfcf208aff17f87cc1b00"},"schema_version":"1.0","source":{"id":"2603.22056","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2603.22056","created_at":"2026-05-20T00:04:29Z"},{"alias_kind":"arxiv_version","alias_value":"2603.22056v2","created_at":"2026-05-20T00:04:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.22056","created_at":"2026-05-20T00:04:29Z"},{"alias_kind":"pith_short_12","alias_value":"ILYMW5VHJOXJ","created_at":"2026-05-20T00:04:29Z"},{"alias_kind":"pith_short_16","alias_value":"ILYMW5VHJOXJD6GT","created_at":"2026-05-20T00:04:29Z"},{"alias_kind":"pith_short_8","alias_value":"ILYMW5VH","created_at":"2026-05-20T00:04:29Z"}],"graph_snapshots":[{"event_id":"sha256:6d767527ff35fe8016735807375de816cefc6681d88d8626b320103bc63ce0ee","target":"graph","created_at":"2026-05-20T00:04:29Z","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/2603.22056/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) achieve state-of-the-art (SOTA) performance across language tasks, but are costly to deploy due to their size and resource demands. Knowledge Distillation (KD) addresses this by training smaller Student models to mimic larger Teacher models, improving efficiency without significant performance loss. Dual-Space Knowledge Distillation with Cross-Model Attention (DSKD-CMA) has emerged as a SOTA method for KD between LLMs with distinct tokenizers, yet its internal workings remain largely opaque. In this work, we systematically analyse the attention mechanism of DSKD-CM","authors_text":"Cong-Thanh Do, Kate Knill, Stella Eva Tsiapali","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2026-03-23T14:53:55Z","title":"Dual-Space Knowledge Distillation with Key-Query Matching for Large Language Models with Vocabulary Mismatch"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.22056","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":"sha256:3118186f1d12baf1a20f521f29d1f1d1698936fb130ea5c7317fd5d61a0f2a13","target":"record","created_at":"2026-05-20T00:04:29Z","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":"dad5156d43b04df52db3451cf11332d9868c3f58a0b8e15dee5c7d694e7e24c9","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2026-03-23T14:53:55Z","title_canon_sha256":"0097e9376adcf1b7fdf3d3742de173b12760ef9d4d6cfcf208aff17f87cc1b00"},"schema_version":"1.0","source":{"id":"2603.22056","kind":"arxiv","version":2}},"canonical_sha256":"42f0cb76a74bae91f8d321ed5ebd0920027dc9c07599799d843a418a69b7d381","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"42f0cb76a74bae91f8d321ed5ebd0920027dc9c07599799d843a418a69b7d381","first_computed_at":"2026-05-20T00:04:29.139895Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:04:29.139895Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"wVgkawp3R+3xmZxqrN9EFZmV99Qgymln/NdrrKAFztTRHL1EVZcvh5EEHUR0a3NTF98jMCWYfaYw9VIBMxPUCw==","signature_status":"signed_v1","signed_at":"2026-05-20T00:04:29.140986Z","signed_message":"canonical_sha256_bytes"},"source_id":"2603.22056","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3118186f1d12baf1a20f521f29d1f1d1698936fb130ea5c7317fd5d61a0f2a13","sha256:6d767527ff35fe8016735807375de816cefc6681d88d8626b320103bc63ce0ee"],"state_sha256":"acdac3d2b7674cc27c40dccad60294e80334e92efc3867605a5b3e3bc0280147"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"XIOXhnDnMyeWCwmS51AuXX3C6QMbTfszp7d57K2XIsMDujdePAz+if8lH3+dkt7GnVwaEZA8jeoRyKFQw1pwBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-26T06:35:26.332134Z","bundle_sha256":"b40d23adff3ee53c790c43f59ba8dbf06af3c13eb9cf31a765009b33b48afc8f"}}