{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:Q7LEFODESOZBVRMPU2TDGYPA7I","short_pith_number":"pith:Q7LEFODE","schema_version":"1.0","canonical_sha256":"87d642b86493b21ac58fa6a63361e0fa0bd054f6096253973f0357708b8d6fe7","source":{"kind":"arxiv","id":"2403.18159","version":2},"attestation_state":"computed","paper":{"title":"Oh! We Freeze: Improving Quantized Knowledge Distillation via Signal Propagation Analysis for Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Harris Teague, Jun Ma, Kartikeya Bhardwaj, Kyunggeun Lee, Nilesh Prasad Pandey, Sweta Priyadarshi","submitted_at":"2024-03-26T23:51:44Z","abstract_excerpt":"Large generative models such as large language models (LLMs) and diffusion models have revolutionized the fields of NLP and computer vision respectively. However, their slow inference, high computation and memory requirement makes it challenging to deploy them on edge devices. In this study, we propose a light-weight quantization aware fine tuning technique using knowledge distillation (KD-QAT) to improve the performance of 4-bit weight quantized LLMs using commonly available datasets to realize a popular language use case, on device chat applications. To improve this paradigm of finetuning, a"},"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":"2403.18159","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-03-26T23:51:44Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"0f2212e80de93013c01c1aaa1237f30562241eed61c5e62264906190f9b035ea","abstract_canon_sha256":"c1c2fca85fbea3ef1c284a899181cde257981b942a16db7c50c9c743d3a8e879"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:01:39.706377Z","signature_b64":"RbvQIsgm+ujxX02sFRrqv0OLQKmWhlI00KocOCc5gTu+RBiGhJ1wHQEXYr+lfHe7hALZmWtBlKq3A9kC+UWtCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"87d642b86493b21ac58fa6a63361e0fa0bd054f6096253973f0357708b8d6fe7","last_reissued_at":"2026-07-05T08:01:39.705856Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:01:39.705856Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Oh! We Freeze: Improving Quantized Knowledge Distillation via Signal Propagation Analysis for Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Harris Teague, Jun Ma, Kartikeya Bhardwaj, Kyunggeun Lee, Nilesh Prasad Pandey, Sweta Priyadarshi","submitted_at":"2024-03-26T23:51:44Z","abstract_excerpt":"Large generative models such as large language models (LLMs) and diffusion models have revolutionized the fields of NLP and computer vision respectively. However, their slow inference, high computation and memory requirement makes it challenging to deploy them on edge devices. In this study, we propose a light-weight quantization aware fine tuning technique using knowledge distillation (KD-QAT) to improve the performance of 4-bit weight quantized LLMs using commonly available datasets to realize a popular language use case, on device chat applications. To improve this paradigm of finetuning, a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.18159","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/2403.18159/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":"2403.18159","created_at":"2026-07-05T08:01:39.705912+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.18159v2","created_at":"2026-07-05T08:01:39.705912+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.18159","created_at":"2026-07-05T08:01:39.705912+00:00"},{"alias_kind":"pith_short_12","alias_value":"Q7LEFODESOZB","created_at":"2026-07-05T08:01:39.705912+00:00"},{"alias_kind":"pith_short_16","alias_value":"Q7LEFODESOZBVRMP","created_at":"2026-07-05T08:01:39.705912+00:00"},{"alias_kind":"pith_short_8","alias_value":"Q7LEFODE","created_at":"2026-07-05T08:01:39.705912+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/Q7LEFODESOZBVRMPU2TDGYPA7I","json":"https://pith.science/pith/Q7LEFODESOZBVRMPU2TDGYPA7I.json","graph_json":"https://pith.science/api/pith-number/Q7LEFODESOZBVRMPU2TDGYPA7I/graph.json","events_json":"https://pith.science/api/pith-number/Q7LEFODESOZBVRMPU2TDGYPA7I/events.json","paper":"https://pith.science/paper/Q7LEFODE"},"agent_actions":{"view_html":"https://pith.science/pith/Q7LEFODESOZBVRMPU2TDGYPA7I","download_json":"https://pith.science/pith/Q7LEFODESOZBVRMPU2TDGYPA7I.json","view_paper":"https://pith.science/paper/Q7LEFODE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.18159&json=true","fetch_graph":"https://pith.science/api/pith-number/Q7LEFODESOZBVRMPU2TDGYPA7I/graph.json","fetch_events":"https://pith.science/api/pith-number/Q7LEFODESOZBVRMPU2TDGYPA7I/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Q7LEFODESOZBVRMPU2TDGYPA7I/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Q7LEFODESOZBVRMPU2TDGYPA7I/action/storage_attestation","attest_author":"https://pith.science/pith/Q7LEFODESOZBVRMPU2TDGYPA7I/action/author_attestation","sign_citation":"https://pith.science/pith/Q7LEFODESOZBVRMPU2TDGYPA7I/action/citation_signature","submit_replication":"https://pith.science/pith/Q7LEFODESOZBVRMPU2TDGYPA7I/action/replication_record"}},"created_at":"2026-07-05T08:01:39.705912+00:00","updated_at":"2026-07-05T08:01:39.705912+00:00"}