{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:26I5ZQJGYFZMM2EBBPY47D7XFJ","short_pith_number":"pith:26I5ZQJG","schema_version":"1.0","canonical_sha256":"d791dcc126c172c668810bf1cf8ff72a51c7b58d5c51ee2c1bbc4996c46e599a","source":{"kind":"arxiv","id":"2606.22942","version":1},"attestation_state":"computed","paper":{"title":"Understanding Knowledge Distillation in Post-Training: When It Helps and When It Fails","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Haoyun Deng, Kaiqiang Song, Lu Wang, Sathish Reddy Indurthi, Shujian Liu, Simin Ma, Song Wang, Xin Liu","submitted_at":"2026-06-22T07:19:43Z","abstract_excerpt":"Large language models (LLMs) achieve strong performance across many tasks, but their high computational cost limits deployment in resource-constrained environments. Knowledge Distillation (KD) offers a practical solution by transferring knowledge from a teacher model of a larger size to a smaller student model. While prior work has mainly examined task-specific or small-scale settings, the post-training stage for building general instruction-following models has received limited attention. In this paper, we conduct a systematic study of KD in post-training using the large-scale Tulu 3 dataset."},"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":"2606.22942","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-22T07:19:43Z","cross_cats_sorted":[],"title_canon_sha256":"1daaa27ebc80bcd1fc234f89676b0e2a6442ed6c0c57b8a71ab4834ac1021595","abstract_canon_sha256":"d9519c426bc591d5af1dc857ac43f6563d21d6a17556be7806d3c487b48dfdeb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T03:14:04.883304Z","signature_b64":"E98d11eWd0CKrvpF6bUUFzX5L13qAYjI1YcEAkyKmqwTg8QzLTpj3IWoc8uLODXZlRgVtGEgUcnJdIZ6dKTWAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d791dcc126c172c668810bf1cf8ff72a51c7b58d5c51ee2c1bbc4996c46e599a","last_reissued_at":"2026-06-23T03:14:04.882901Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T03:14:04.882901Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Understanding Knowledge Distillation in Post-Training: When It Helps and When It Fails","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Haoyun Deng, Kaiqiang Song, Lu Wang, Sathish Reddy Indurthi, Shujian Liu, Simin Ma, Song Wang, Xin Liu","submitted_at":"2026-06-22T07:19:43Z","abstract_excerpt":"Large language models (LLMs) achieve strong performance across many tasks, but their high computational cost limits deployment in resource-constrained environments. Knowledge Distillation (KD) offers a practical solution by transferring knowledge from a teacher model of a larger size to a smaller student model. While prior work has mainly examined task-specific or small-scale settings, the post-training stage for building general instruction-following models has received limited attention. In this paper, we conduct a systematic study of KD in post-training using the large-scale Tulu 3 dataset."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.22942","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/2606.22942/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":"2606.22942","created_at":"2026-06-23T03:14:04.882960+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.22942v1","created_at":"2026-06-23T03:14:04.882960+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.22942","created_at":"2026-06-23T03:14:04.882960+00:00"},{"alias_kind":"pith_short_12","alias_value":"26I5ZQJGYFZM","created_at":"2026-06-23T03:14:04.882960+00:00"},{"alias_kind":"pith_short_16","alias_value":"26I5ZQJGYFZMM2EB","created_at":"2026-06-23T03:14:04.882960+00:00"},{"alias_kind":"pith_short_8","alias_value":"26I5ZQJG","created_at":"2026-06-23T03:14:04.882960+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/26I5ZQJGYFZMM2EBBPY47D7XFJ","json":"https://pith.science/pith/26I5ZQJGYFZMM2EBBPY47D7XFJ.json","graph_json":"https://pith.science/api/pith-number/26I5ZQJGYFZMM2EBBPY47D7XFJ/graph.json","events_json":"https://pith.science/api/pith-number/26I5ZQJGYFZMM2EBBPY47D7XFJ/events.json","paper":"https://pith.science/paper/26I5ZQJG"},"agent_actions":{"view_html":"https://pith.science/pith/26I5ZQJGYFZMM2EBBPY47D7XFJ","download_json":"https://pith.science/pith/26I5ZQJGYFZMM2EBBPY47D7XFJ.json","view_paper":"https://pith.science/paper/26I5ZQJG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.22942&json=true","fetch_graph":"https://pith.science/api/pith-number/26I5ZQJGYFZMM2EBBPY47D7XFJ/graph.json","fetch_events":"https://pith.science/api/pith-number/26I5ZQJGYFZMM2EBBPY47D7XFJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/26I5ZQJGYFZMM2EBBPY47D7XFJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/26I5ZQJGYFZMM2EBBPY47D7XFJ/action/storage_attestation","attest_author":"https://pith.science/pith/26I5ZQJGYFZMM2EBBPY47D7XFJ/action/author_attestation","sign_citation":"https://pith.science/pith/26I5ZQJGYFZMM2EBBPY47D7XFJ/action/citation_signature","submit_replication":"https://pith.science/pith/26I5ZQJGYFZMM2EBBPY47D7XFJ/action/replication_record"}},"created_at":"2026-06-23T03:14:04.882960+00:00","updated_at":"2026-06-23T03:14:04.882960+00:00"}