{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:2UVJMP4O2M3DUYVRISAWJJHSUE","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":"70b889e21366465fea4eb846d654c26ab1d9bdfa07734ba457d8be5ded606249","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2025-03-21T05:58:18Z","title_canon_sha256":"13ea0ad45d71e1cfbae70c3f6403a32dd6d4b0d341b214d72b6adc424d1e7981"},"schema_version":"1.0","source":{"id":"2503.16870","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.16870","created_at":"2026-07-05T11:42:27Z"},{"alias_kind":"arxiv_version","alias_value":"2503.16870v2","created_at":"2026-07-05T11:42:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.16870","created_at":"2026-07-05T11:42:27Z"},{"alias_kind":"pith_short_12","alias_value":"2UVJMP4O2M3D","created_at":"2026-07-05T11:42:27Z"},{"alias_kind":"pith_short_16","alias_value":"2UVJMP4O2M3DUYVR","created_at":"2026-07-05T11:42:27Z"},{"alias_kind":"pith_short_8","alias_value":"2UVJMP4O","created_at":"2026-07-05T11:42:27Z"}],"graph_snapshots":[{"event_id":"sha256:b8efed69b718a11cfb969f6aebe472c3fb725c7f976479ba758ca8be00b5b4df","target":"graph","created_at":"2026-07-05T11:42:27Z","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/2503.16870/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Knowledge distillation can be a cost-effective technique to distill knowledge in Large Language Models, if the teacher output logits can be pre-computed and cached. However, successfully applying this to pre-training remains largely unexplored. In this work, we prove that naive approaches for sparse knowledge distillation such as caching Top-K probabilities, while intuitive, provide biased estimates of teacher probability distribution to the student, resulting in suboptimal performance and calibration. We propose an importance-sampling-based method `Random Sampling Knowledge Distillation', whi","authors_text":"Akhil Kedia, Anshumann, Haejun Lee, Jinwoo Ahn, Joohyung Lee, Kangwook Lee, Mohd Abbas Zaidi, Taehwak Kwon","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2025-03-21T05:58:18Z","title":"Sparse Logit Sampling: Accelerating Knowledge Distillation in LLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.16870","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":"sha256:1c8b7d67ccf2062c627b8ebe30e55ea64d2c24091df00b43b67539d494c98a23","target":"record","created_at":"2026-07-05T11:42:27Z","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":"70b889e21366465fea4eb846d654c26ab1d9bdfa07734ba457d8be5ded606249","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2025-03-21T05:58:18Z","title_canon_sha256":"13ea0ad45d71e1cfbae70c3f6403a32dd6d4b0d341b214d72b6adc424d1e7981"},"schema_version":"1.0","source":{"id":"2503.16870","kind":"arxiv","version":2}},"canonical_sha256":"d52a963f8ed3363a62b1448164a4f2a10b0f012c904161cf987bf3c16190e01f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d52a963f8ed3363a62b1448164a4f2a10b0f012c904161cf987bf3c16190e01f","first_computed_at":"2026-07-05T11:42:27.673323Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:42:27.673323Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"n2hIQwlDLiCf6wRESkHFYSKReOvlMZkkdOYxvxhhehqT7eYTCWZpYSrpt0Ayx9LUFBhjcaRZSjOSSey1l8uKBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:42:27.673872Z","signed_message":"canonical_sha256_bytes"},"source_id":"2503.16870","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:1c8b7d67ccf2062c627b8ebe30e55ea64d2c24091df00b43b67539d494c98a23","sha256:b8efed69b718a11cfb969f6aebe472c3fb725c7f976479ba758ca8be00b5b4df"],"state_sha256":"f019ae00eb2419670fabe3a83c9d5eadedebc412358089bdbaf0d21c584053b5"}