{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:TSN233DVG2QDINBLWCFMGNJEEZ","short_pith_number":"pith:TSN233DV","schema_version":"1.0","canonical_sha256":"9c9badec7536a034342bb08ac33524266aca128504470e6bd6d0327f29c3a528","source":{"kind":"arxiv","id":"2606.21851","version":1},"attestation_state":"computed","paper":{"title":"TALAS: Teacher-Anchored Layer Alignment with Adaptive Sharpness-Aware Minimization for Embedding Distillation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Hoang Son Nguyen, Linh Ngo Van, Nguyen Thi Ngoc Diep, Pham Khanh Chi, Quoc Phong Dao, Thien Huu Nguyen, Trung Le","submitted_at":"2026-06-20T03:17:29Z","abstract_excerpt":"Knowledge Distillation (KD) has established itself as a pivotal technique for compressing large pre-trained language models. However, existing methods that force a student to strictly mimic the teacher's sentence embeddings or internal features often incur prohibitive computational costs and yield suboptimal performance due to the inherent capacity gap. To address these challenges, we propose TALAS (Teacher-Anchored Layer Alignment with Sharpness-aware minimization), a unified framework that synergizes hierarchical (multi-layer) alignment with robust optimization. First, we introduce a Teacher"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2606.21851","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-20T03:17:29Z","cross_cats_sorted":[],"title_canon_sha256":"fb87632e2e9c4db20f95ce0674e80971db6b4024855e0c465cd653fe8dadcddf","abstract_canon_sha256":"a6abe71900aec75f5a165b342b6082112f6463acac096020d74add307912bcba"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T02:13:00.807776Z","signature_b64":"7d4f7vK5kNx1Jq7riag2JR+ltXpExy47uyWJk+TclgHK73jsttVdRxVl2nRyMUqUF1jxSp+cjSiVdMVNLN/IBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9c9badec7536a034342bb08ac33524266aca128504470e6bd6d0327f29c3a528","last_reissued_at":"2026-06-23T02:13:00.807398Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T02:13:00.807398Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TALAS: Teacher-Anchored Layer Alignment with Adaptive Sharpness-Aware Minimization for Embedding Distillation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Hoang Son Nguyen, Linh Ngo Van, Nguyen Thi Ngoc Diep, Pham Khanh Chi, Quoc Phong Dao, Thien Huu Nguyen, Trung Le","submitted_at":"2026-06-20T03:17:29Z","abstract_excerpt":"Knowledge Distillation (KD) has established itself as a pivotal technique for compressing large pre-trained language models. However, existing methods that force a student to strictly mimic the teacher's sentence embeddings or internal features often incur prohibitive computational costs and yield suboptimal performance due to the inherent capacity gap. To address these challenges, we propose TALAS (Teacher-Anchored Layer Alignment with Sharpness-aware minimization), a unified framework that synergizes hierarchical (multi-layer) alignment with robust optimization. First, we introduce a Teacher"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.21851","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/2606.21851/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2606.21851","created_at":"2026-06-23T02:13:00.807460+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.21851v1","created_at":"2026-06-23T02:13:00.807460+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.21851","created_at":"2026-06-23T02:13:00.807460+00:00"},{"alias_kind":"pith_short_12","alias_value":"TSN233DVG2QD","created_at":"2026-06-23T02:13:00.807460+00:00"},{"alias_kind":"pith_short_16","alias_value":"TSN233DVG2QDINBL","created_at":"2026-06-23T02:13:00.807460+00:00"},{"alias_kind":"pith_short_8","alias_value":"TSN233DV","created_at":"2026-06-23T02:13:00.807460+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TSN233DVG2QDINBLWCFMGNJEEZ","json":"https://pith.science/pith/TSN233DVG2QDINBLWCFMGNJEEZ.json","graph_json":"https://pith.science/api/pith-number/TSN233DVG2QDINBLWCFMGNJEEZ/graph.json","events_json":"https://pith.science/api/pith-number/TSN233DVG2QDINBLWCFMGNJEEZ/events.json","paper":"https://pith.science/paper/TSN233DV"},"agent_actions":{"view_html":"https://pith.science/pith/TSN233DVG2QDINBLWCFMGNJEEZ","download_json":"https://pith.science/pith/TSN233DVG2QDINBLWCFMGNJEEZ.json","view_paper":"https://pith.science/paper/TSN233DV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.21851&json=true","fetch_graph":"https://pith.science/api/pith-number/TSN233DVG2QDINBLWCFMGNJEEZ/graph.json","fetch_events":"https://pith.science/api/pith-number/TSN233DVG2QDINBLWCFMGNJEEZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TSN233DVG2QDINBLWCFMGNJEEZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TSN233DVG2QDINBLWCFMGNJEEZ/action/storage_attestation","attest_author":"https://pith.science/pith/TSN233DVG2QDINBLWCFMGNJEEZ/action/author_attestation","sign_citation":"https://pith.science/pith/TSN233DVG2QDINBLWCFMGNJEEZ/action/citation_signature","submit_replication":"https://pith.science/pith/TSN233DVG2QDINBLWCFMGNJEEZ/action/replication_record"}},"created_at":"2026-06-23T02:13:00.807460+00:00","updated_at":"2026-06-23T02:13:00.807460+00:00"}