{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:7O6P5JKEGW7R5WOA6WAT5DM2L2","short_pith_number":"pith:7O6P5JKE","schema_version":"1.0","canonical_sha256":"fbbcfea54435bf1ed9c0f5813e8d9a5eaa46d5082e1e6e38a6b7823ab8bff185","source":{"kind":"arxiv","id":"2507.07630","version":1},"attestation_state":"computed","paper":{"title":"Exploring the Limits of Model Compression in LLMs: A Knowledge Distillation Study on QA Tasks","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Joyeeta Datta, Niclas Doll, Qusai Ramadan, Zeyd Boukhers","submitted_at":"2025-07-10T10:54:05Z","abstract_excerpt":"Large Language Models (LLMs) have demonstrated outstanding performance across a range of NLP tasks, however, their computational demands hinder their deployment in real-world, resource-constrained environments. This work investigates the extent to which LLMs can be compressed using Knowledge Distillation (KD) while maintaining strong performance on Question Answering (QA) tasks. We evaluate student models distilled from the Pythia and Qwen2.5 families on two QA benchmarks, SQuAD and MLQA, under zero-shot and one-shot prompting conditions. Results show that student models retain over 90% of the"},"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":"2507.07630","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-10T10:54:05Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"8ef7766d925ea802a4d172305f24cc7ac1fcc8bb92d1e97c66eadc46b0334dd0","abstract_canon_sha256":"d896868ca7c6a0800bb9ca2bcd1691f31e917b36543264928a9caa389a589f2b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:35:02.123088Z","signature_b64":"KJjyHGDfZkPZRmy9jl5alR0qeJM5rbYVzMF6xUVHLsoelapXgbAQgBdIVkKnxJp8IQq6G+OVIAYhZnKAM2foBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fbbcfea54435bf1ed9c0f5813e8d9a5eaa46d5082e1e6e38a6b7823ab8bff185","last_reissued_at":"2026-07-05T11:35:02.122653Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:35:02.122653Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Exploring the Limits of Model Compression in LLMs: A Knowledge Distillation Study on QA Tasks","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Joyeeta Datta, Niclas Doll, Qusai Ramadan, Zeyd Boukhers","submitted_at":"2025-07-10T10:54:05Z","abstract_excerpt":"Large Language Models (LLMs) have demonstrated outstanding performance across a range of NLP tasks, however, their computational demands hinder their deployment in real-world, resource-constrained environments. This work investigates the extent to which LLMs can be compressed using Knowledge Distillation (KD) while maintaining strong performance on Question Answering (QA) tasks. We evaluate student models distilled from the Pythia and Qwen2.5 families on two QA benchmarks, SQuAD and MLQA, under zero-shot and one-shot prompting conditions. Results show that student models retain over 90% of the"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.07630","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/2507.07630/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":"2507.07630","created_at":"2026-07-05T11:35:02.122712+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.07630v1","created_at":"2026-07-05T11:35:02.122712+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.07630","created_at":"2026-07-05T11:35:02.122712+00:00"},{"alias_kind":"pith_short_12","alias_value":"7O6P5JKEGW7R","created_at":"2026-07-05T11:35:02.122712+00:00"},{"alias_kind":"pith_short_16","alias_value":"7O6P5JKEGW7R5WOA","created_at":"2026-07-05T11:35:02.122712+00:00"},{"alias_kind":"pith_short_8","alias_value":"7O6P5JKE","created_at":"2026-07-05T11:35:02.122712+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/7O6P5JKEGW7R5WOA6WAT5DM2L2","json":"https://pith.science/pith/7O6P5JKEGW7R5WOA6WAT5DM2L2.json","graph_json":"https://pith.science/api/pith-number/7O6P5JKEGW7R5WOA6WAT5DM2L2/graph.json","events_json":"https://pith.science/api/pith-number/7O6P5JKEGW7R5WOA6WAT5DM2L2/events.json","paper":"https://pith.science/paper/7O6P5JKE"},"agent_actions":{"view_html":"https://pith.science/pith/7O6P5JKEGW7R5WOA6WAT5DM2L2","download_json":"https://pith.science/pith/7O6P5JKEGW7R5WOA6WAT5DM2L2.json","view_paper":"https://pith.science/paper/7O6P5JKE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.07630&json=true","fetch_graph":"https://pith.science/api/pith-number/7O6P5JKEGW7R5WOA6WAT5DM2L2/graph.json","fetch_events":"https://pith.science/api/pith-number/7O6P5JKEGW7R5WOA6WAT5DM2L2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7O6P5JKEGW7R5WOA6WAT5DM2L2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7O6P5JKEGW7R5WOA6WAT5DM2L2/action/storage_attestation","attest_author":"https://pith.science/pith/7O6P5JKEGW7R5WOA6WAT5DM2L2/action/author_attestation","sign_citation":"https://pith.science/pith/7O6P5JKEGW7R5WOA6WAT5DM2L2/action/citation_signature","submit_replication":"https://pith.science/pith/7O6P5JKEGW7R5WOA6WAT5DM2L2/action/replication_record"}},"created_at":"2026-07-05T11:35:02.122712+00:00","updated_at":"2026-07-05T11:35:02.122712+00:00"}