{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:2TN2AYSGXUT2B3S3UUCYY2VCXD","short_pith_number":"pith:2TN2AYSG","schema_version":"1.0","canonical_sha256":"d4dba06246bd27a0ee5ba5058c6aa2b8f42c593d6e198c5e69681dad7a26b5d7","source":{"kind":"arxiv","id":"2411.05046","version":1},"attestation_state":"computed","paper":{"title":"PhoneLM:an Efficient and Capable Small Language Model Family through Principled Pre-training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Ao Zhou, Chenghua Wang, Mengwei Xu, Rongjie Yi, Shangguang Wang, Weikai Xie, Xiang Li, Xiwen Zhang, Zhenyan Lu","submitted_at":"2024-11-07T02:19:00Z","abstract_excerpt":"The interest in developing small language models (SLM) for on-device deployment is fast growing. However, the existing SLM design hardly considers the device hardware characteristics. Instead, this work presents a simple yet effective principle for SLM design: architecture searching for (near-)optimal runtime efficiency before pre-training. Guided by this principle, we develop PhoneLM SLM family (currently with 0.5B and 1.5B versions), that acheive the state-of-the-art capability-efficiency tradeoff among those with similar parameter size. We fully open-source the code, weights, and training d"},"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":"2411.05046","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-11-07T02:19:00Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"402ca83944f91106c1b1f155a4e888c25ef1651d23fda603ce5ef98150e79103","abstract_canon_sha256":"24e44734b2c3d3ebcd1deb1ce69c0337bd28d193d2e84e3d2f9258fa98771e2e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:32:51.878520Z","signature_b64":"aKiyaX//xPaKptoWz9Y9RNOWCQNHLJG5Au7Eyn4U1FWDyLqOnPmp1gtaFft4J+woetfEWqlHpCoGmrpPDgOmAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d4dba06246bd27a0ee5ba5058c6aa2b8f42c593d6e198c5e69681dad7a26b5d7","last_reissued_at":"2026-07-05T09:32:51.877969Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:32:51.877969Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"PhoneLM:an Efficient and Capable Small Language Model Family through Principled Pre-training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Ao Zhou, Chenghua Wang, Mengwei Xu, Rongjie Yi, Shangguang Wang, Weikai Xie, Xiang Li, Xiwen Zhang, Zhenyan Lu","submitted_at":"2024-11-07T02:19:00Z","abstract_excerpt":"The interest in developing small language models (SLM) for on-device deployment is fast growing. However, the existing SLM design hardly considers the device hardware characteristics. Instead, this work presents a simple yet effective principle for SLM design: architecture searching for (near-)optimal runtime efficiency before pre-training. Guided by this principle, we develop PhoneLM SLM family (currently with 0.5B and 1.5B versions), that acheive the state-of-the-art capability-efficiency tradeoff among those with similar parameter size. We fully open-source the code, weights, and training d"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.05046","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/2411.05046/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":"2411.05046","created_at":"2026-07-05T09:32:51.878036+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.05046v1","created_at":"2026-07-05T09:32:51.878036+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.05046","created_at":"2026-07-05T09:32:51.878036+00:00"},{"alias_kind":"pith_short_12","alias_value":"2TN2AYSGXUT2","created_at":"2026-07-05T09:32:51.878036+00:00"},{"alias_kind":"pith_short_16","alias_value":"2TN2AYSGXUT2B3S3","created_at":"2026-07-05T09:32:51.878036+00:00"},{"alias_kind":"pith_short_8","alias_value":"2TN2AYSG","created_at":"2026-07-05T09:32:51.878036+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2503.08223","citing_title":"Will LLMs Scaling Hit the Wall? Breaking Barriers via Distributed Resources on Massive Edge Devices","ref_index":137,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2TN2AYSGXUT2B3S3UUCYY2VCXD","json":"https://pith.science/pith/2TN2AYSGXUT2B3S3UUCYY2VCXD.json","graph_json":"https://pith.science/api/pith-number/2TN2AYSGXUT2B3S3UUCYY2VCXD/graph.json","events_json":"https://pith.science/api/pith-number/2TN2AYSGXUT2B3S3UUCYY2VCXD/events.json","paper":"https://pith.science/paper/2TN2AYSG"},"agent_actions":{"view_html":"https://pith.science/pith/2TN2AYSGXUT2B3S3UUCYY2VCXD","download_json":"https://pith.science/pith/2TN2AYSGXUT2B3S3UUCYY2VCXD.json","view_paper":"https://pith.science/paper/2TN2AYSG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.05046&json=true","fetch_graph":"https://pith.science/api/pith-number/2TN2AYSGXUT2B3S3UUCYY2VCXD/graph.json","fetch_events":"https://pith.science/api/pith-number/2TN2AYSGXUT2B3S3UUCYY2VCXD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2TN2AYSGXUT2B3S3UUCYY2VCXD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2TN2AYSGXUT2B3S3UUCYY2VCXD/action/storage_attestation","attest_author":"https://pith.science/pith/2TN2AYSGXUT2B3S3UUCYY2VCXD/action/author_attestation","sign_citation":"https://pith.science/pith/2TN2AYSGXUT2B3S3UUCYY2VCXD/action/citation_signature","submit_replication":"https://pith.science/pith/2TN2AYSGXUT2B3S3UUCYY2VCXD/action/replication_record"}},"created_at":"2026-07-05T09:32:51.878036+00:00","updated_at":"2026-07-05T09:32:51.878036+00:00"}