{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:JOTSOH4J7HBDSIOJ257OU4MDSJ","short_pith_number":"pith:JOTSOH4J","schema_version":"1.0","canonical_sha256":"4ba7271f89f9c23921c9d77eea7183924333dff9b0e8200f2ff02d279600a291","source":{"kind":"arxiv","id":"2506.03510","version":1},"attestation_state":"computed","paper":{"title":"Accurate Sublayer Pruning for Large Language Models by Exploiting Latency and Tunability Information","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Hyunjik Jo, Jinsik Lee, JongJin Kim, Seungcheol Park, Sojin Lee, U Kang","submitted_at":"2025-06-04T02:53:34Z","abstract_excerpt":"How can we accelerate large language models(LLMs) without sacrificing accuracy? The slow inference speed of LLMs hinders us to benefit from their remarkable performance in diverse applications. This is mainly because numerous sublayers are stacked together in LLMs. Sublayer pruning compresses and expedites LLMs via removing unnecessary sublayers. However, existing sublayer pruning algorithms are limited in accuracy since they naively select sublayers to prune, overlooking the different characteristics of each sublayer. In this paper, we propose SPRINT (Sublayer PRuning wIth LateNcy and Tunabil"},"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":"2506.03510","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-04T02:53:34Z","cross_cats_sorted":[],"title_canon_sha256":"b641fb5f27d8cd21d59c7853ecab2650ed789c526cbab72a26225d12a3fda311","abstract_canon_sha256":"c79d7957d7ae0b2a7bb3ca62bd55e2da2d293a8eb0e2e242ea219f65a30d14ae"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:15:28.990045Z","signature_b64":"CT0xyUia90bru6hgiUcjwMU0Pd6Hnj3t5wZo6FHitO0i/noudNB4sHcVCUX+I3haBmHMwAHNQFgxTU6O5yVpCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4ba7271f89f9c23921c9d77eea7183924333dff9b0e8200f2ff02d279600a291","last_reissued_at":"2026-07-05T11:15:28.989638Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:15:28.989638Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Accurate Sublayer Pruning for Large Language Models by Exploiting Latency and Tunability Information","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Hyunjik Jo, Jinsik Lee, JongJin Kim, Seungcheol Park, Sojin Lee, U Kang","submitted_at":"2025-06-04T02:53:34Z","abstract_excerpt":"How can we accelerate large language models(LLMs) without sacrificing accuracy? The slow inference speed of LLMs hinders us to benefit from their remarkable performance in diverse applications. This is mainly because numerous sublayers are stacked together in LLMs. Sublayer pruning compresses and expedites LLMs via removing unnecessary sublayers. However, existing sublayer pruning algorithms are limited in accuracy since they naively select sublayers to prune, overlooking the different characteristics of each sublayer. In this paper, we propose SPRINT (Sublayer PRuning wIth LateNcy and Tunabil"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.03510","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/2506.03510/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":"2506.03510","created_at":"2026-07-05T11:15:28.989696+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.03510v1","created_at":"2026-07-05T11:15:28.989696+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.03510","created_at":"2026-07-05T11:15:28.989696+00:00"},{"alias_kind":"pith_short_12","alias_value":"JOTSOH4J7HBD","created_at":"2026-07-05T11:15:28.989696+00:00"},{"alias_kind":"pith_short_16","alias_value":"JOTSOH4J7HBDSIOJ","created_at":"2026-07-05T11:15:28.989696+00:00"},{"alias_kind":"pith_short_8","alias_value":"JOTSOH4J","created_at":"2026-07-05T11:15:28.989696+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/JOTSOH4J7HBDSIOJ257OU4MDSJ","json":"https://pith.science/pith/JOTSOH4J7HBDSIOJ257OU4MDSJ.json","graph_json":"https://pith.science/api/pith-number/JOTSOH4J7HBDSIOJ257OU4MDSJ/graph.json","events_json":"https://pith.science/api/pith-number/JOTSOH4J7HBDSIOJ257OU4MDSJ/events.json","paper":"https://pith.science/paper/JOTSOH4J"},"agent_actions":{"view_html":"https://pith.science/pith/JOTSOH4J7HBDSIOJ257OU4MDSJ","download_json":"https://pith.science/pith/JOTSOH4J7HBDSIOJ257OU4MDSJ.json","view_paper":"https://pith.science/paper/JOTSOH4J","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.03510&json=true","fetch_graph":"https://pith.science/api/pith-number/JOTSOH4J7HBDSIOJ257OU4MDSJ/graph.json","fetch_events":"https://pith.science/api/pith-number/JOTSOH4J7HBDSIOJ257OU4MDSJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JOTSOH4J7HBDSIOJ257OU4MDSJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JOTSOH4J7HBDSIOJ257OU4MDSJ/action/storage_attestation","attest_author":"https://pith.science/pith/JOTSOH4J7HBDSIOJ257OU4MDSJ/action/author_attestation","sign_citation":"https://pith.science/pith/JOTSOH4J7HBDSIOJ257OU4MDSJ/action/citation_signature","submit_replication":"https://pith.science/pith/JOTSOH4J7HBDSIOJ257OU4MDSJ/action/replication_record"}},"created_at":"2026-07-05T11:15:28.989696+00:00","updated_at":"2026-07-05T11:15:28.989696+00:00"}