{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:3FVA3U33ZHCBNZB6B3CTKNVFSE","short_pith_number":"pith:3FVA3U33","schema_version":"1.0","canonical_sha256":"d96a0dd37bc9c416e43e0ec53536a5913680471776e1e3529212bc29f66975f9","source":{"kind":"arxiv","id":"2402.10787","version":2},"attestation_state":"computed","paper":{"title":"Squat: Quant Small Language Models on the Edge","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Changdi Yang, Chao Wu, Cheng Lyu, Lei Lu, Peiyan Dong, Pu Zhao, Xuan Shen, Yanyue Xie, Yanzhi Wang, Yifan Gong, Zhaoyang Han, Zhenglun Kong","submitted_at":"2024-02-16T16:10:38Z","abstract_excerpt":"A growing trend has emerged in designing high-quality Small Language Models (SLMs) with a few million parameters. This trend is driven by the increasing concerns over cloud costs, privacy, and latency. Considering that full parameter training is feasible for SLMs on mobile devices, Quantization-Aware Training (QAT) is employed to improve efficiency by reducing computational overhead and memory footprint. However, previous QAT works adopt fine-grained quantization methods to compress models with billions of parameters on GPUs, incompatible with current commodity hardware, such as mobile and edg"},"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":"2402.10787","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-02-16T16:10:38Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"30a213d60fad915cbe6148ec7423a94477fe2fa8d41d9f470a908a828c4a49b2","abstract_canon_sha256":"cf87a1341491abfe98cc616f00812093a3af9f7a32a25d1a1022f21bc6881f69"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:30:20.129665Z","signature_b64":"mR17GiDRS3WUhRc6Cfid0VjXWihtlmnwBNpBOm0dUvwqpl1l/1cDlT3L9NxctUEm4RWaQmYT/piGMXNKKUggBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d96a0dd37bc9c416e43e0ec53536a5913680471776e1e3529212bc29f66975f9","last_reissued_at":"2026-07-05T11:30:20.129126Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:30:20.129126Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Squat: Quant Small Language Models on the Edge","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Changdi Yang, Chao Wu, Cheng Lyu, Lei Lu, Peiyan Dong, Pu Zhao, Xuan Shen, Yanyue Xie, Yanzhi Wang, Yifan Gong, Zhaoyang Han, Zhenglun Kong","submitted_at":"2024-02-16T16:10:38Z","abstract_excerpt":"A growing trend has emerged in designing high-quality Small Language Models (SLMs) with a few million parameters. This trend is driven by the increasing concerns over cloud costs, privacy, and latency. Considering that full parameter training is feasible for SLMs on mobile devices, Quantization-Aware Training (QAT) is employed to improve efficiency by reducing computational overhead and memory footprint. However, previous QAT works adopt fine-grained quantization methods to compress models with billions of parameters on GPUs, incompatible with current commodity hardware, such as mobile and edg"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.10787","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/2402.10787/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":"2402.10787","created_at":"2026-07-05T11:30:20.129209+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.10787v2","created_at":"2026-07-05T11:30:20.129209+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.10787","created_at":"2026-07-05T11:30:20.129209+00:00"},{"alias_kind":"pith_short_12","alias_value":"3FVA3U33ZHCB","created_at":"2026-07-05T11:30:20.129209+00:00"},{"alias_kind":"pith_short_16","alias_value":"3FVA3U33ZHCBNZB6","created_at":"2026-07-05T11:30:20.129209+00:00"},{"alias_kind":"pith_short_8","alias_value":"3FVA3U33","created_at":"2026-07-05T11:30:20.129209+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/3FVA3U33ZHCBNZB6B3CTKNVFSE","json":"https://pith.science/pith/3FVA3U33ZHCBNZB6B3CTKNVFSE.json","graph_json":"https://pith.science/api/pith-number/3FVA3U33ZHCBNZB6B3CTKNVFSE/graph.json","events_json":"https://pith.science/api/pith-number/3FVA3U33ZHCBNZB6B3CTKNVFSE/events.json","paper":"https://pith.science/paper/3FVA3U33"},"agent_actions":{"view_html":"https://pith.science/pith/3FVA3U33ZHCBNZB6B3CTKNVFSE","download_json":"https://pith.science/pith/3FVA3U33ZHCBNZB6B3CTKNVFSE.json","view_paper":"https://pith.science/paper/3FVA3U33","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.10787&json=true","fetch_graph":"https://pith.science/api/pith-number/3FVA3U33ZHCBNZB6B3CTKNVFSE/graph.json","fetch_events":"https://pith.science/api/pith-number/3FVA3U33ZHCBNZB6B3CTKNVFSE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3FVA3U33ZHCBNZB6B3CTKNVFSE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3FVA3U33ZHCBNZB6B3CTKNVFSE/action/storage_attestation","attest_author":"https://pith.science/pith/3FVA3U33ZHCBNZB6B3CTKNVFSE/action/author_attestation","sign_citation":"https://pith.science/pith/3FVA3U33ZHCBNZB6B3CTKNVFSE/action/citation_signature","submit_replication":"https://pith.science/pith/3FVA3U33ZHCBNZB6B3CTKNVFSE/action/replication_record"}},"created_at":"2026-07-05T11:30:20.129209+00:00","updated_at":"2026-07-05T11:30:20.129209+00:00"}