{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:DWT2XANHCKBJC72665A4EUJWDL","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":"4a209011ec524ab89114606c2795774c10a891a1e4d1f64f66f42de78d4ce61e","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-02T10:52:49Z","title_canon_sha256":"1b270c6c37c74de47904b5efb3f575514dfda75bb956d0009de8d7354a426b05"},"schema_version":"1.0","source":{"id":"2605.01374","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.01374","created_at":"2026-06-03T01:05:50Z"},{"alias_kind":"arxiv_version","alias_value":"2605.01374v2","created_at":"2026-06-03T01:05:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.01374","created_at":"2026-06-03T01:05:50Z"},{"alias_kind":"pith_short_12","alias_value":"DWT2XANHCKBJ","created_at":"2026-06-03T01:05:50Z"},{"alias_kind":"pith_short_16","alias_value":"DWT2XANHCKBJC726","created_at":"2026-06-03T01:05:50Z"},{"alias_kind":"pith_short_8","alias_value":"DWT2XANH","created_at":"2026-06-03T01:05:50Z"}],"graph_snapshots":[{"event_id":"sha256:1fe73c53fc620547fd043147c973c4ce9905529cb116b4f1de846abb7f61d761","target":"graph","created_at":"2026-06-03T01:05:50Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"MTA consistently outperforms state-of-the-art baselines on standard benchmarks, with ablations confirming the contribution of each component."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That aligning teacher and student representations along their layer-wise transformation trajectory using a layer-adaptive multi-granular strategy (word-level lower, phrase-level higher) will better guide the student to capture the teacher's internal relational structure than fixed-layer or token-level methods."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"MTA improves LLM knowledge distillation by aligning representations along layer-wise trajectories with adaptive granularity from words to phrases using dynamic structural and hidden representation alignment losses."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Multi-granular trajectory alignment improves knowledge distillation by matching teacher and student representations at word level in lower layers and phrase level in higher layers."}],"snapshot_sha256":"054a12ff34e7f9874c0f1e6053c3c6a0dbd8e11508c2f494a714da33f122e632"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-20T18:34:27.146335Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T17:19:42.068043Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2605.01374/integrity.json","findings":[],"snapshot_sha256":"9ae7dd55168262a410c47f3b61079f92c9d350a401bbe530ed675e3582b7d7aa","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Knowledge distillation is a key technique for compressing large language models (LLMs), but most existing methods align representations at fixed layers or token-level outputs, ignoring how representations evolve across depth. As a result, the student is only weakly guided to capture the teacher's internal relational structure during distillation, which limits knowledge transfer. To address this limitation, we propose Multi-Granular Trajectory Alignment (MTA), a framework that aligns teacher and student representations along their layer-wise transformation trajectory. MTA adopts a layer-adaptiv","authors_text":"Linh Ngo Van, Pham Khanh Chi, Quoc Phong Dao, Thanh Hong Nguyen, Thuat Nguyen, Trung Le","cross_cats":[],"headline":"Multi-granular trajectory alignment improves knowledge distillation by matching teacher and student representations at word level in lower layers and phrase level in higher layers.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-02T10:52:49Z","title":"MTA: Multi-Granular Trajectory Alignment for Large Language Model Distillation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.01374","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-09T14:50:59.346561Z","id":"0aa38e82-e6fa-4fdd-9d6e-d325616a4875","model_set":{"reader":"grok-4.3"},"one_line_summary":"MTA improves LLM knowledge distillation by aligning representations along layer-wise trajectories with adaptive granularity from words to phrases using dynamic structural and hidden representation alignment losses.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Multi-granular trajectory alignment improves knowledge distillation by matching teacher and student representations at word level in lower layers and phrase level in higher layers.","strongest_claim":"MTA consistently outperforms state-of-the-art baselines on standard benchmarks, with ablations confirming the contribution of each component.","weakest_assumption":"That aligning teacher and student representations along their layer-wise transformation trajectory using a layer-adaptive multi-granular strategy (word-level lower, phrase-level higher) will better guide the student to capture the teacher's internal relational structure than fixed-layer or token-level methods."}},"verdict_id":"0aa38e82-e6fa-4fdd-9d6e-d325616a4875"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:d54f362144edaf9ab3befc4428e47570e7c35d60e1dd006d367c927ce21f25fd","target":"record","created_at":"2026-06-03T01:05:50Z","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":"4a209011ec524ab89114606c2795774c10a891a1e4d1f64f66f42de78d4ce61e","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-02T10:52:49Z","title_canon_sha256":"1b270c6c37c74de47904b5efb3f575514dfda75bb956d0009de8d7354a426b05"},"schema_version":"1.0","source":{"id":"2605.01374","kind":"arxiv","version":2}},"canonical_sha256":"1da7ab81a71282917f5ef741c251361af80bd16acabf1112ad0d1860bd263913","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1da7ab81a71282917f5ef741c251361af80bd16acabf1112ad0d1860bd263913","first_computed_at":"2026-06-03T01:05:50.830716Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-03T01:05:50.830716Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"RMG0FJ86eN4K5ejLQWhC5FSyy1POhgeL/j8IPUJlWn1OldNOBvhtXoEmIA5+rNoaAeI/6mJRXWduNWqotM75Aw==","signature_status":"signed_v1","signed_at":"2026-06-03T01:05:50.831123Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.01374","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d54f362144edaf9ab3befc4428e47570e7c35d60e1dd006d367c927ce21f25fd","sha256:1fe73c53fc620547fd043147c973c4ce9905529cb116b4f1de846abb7f61d761"],"state_sha256":"c23799783baae3fcabd18ca4ffde471857efe87a329b4004787a2ec1f4833398"}