{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:IJGGTCBPDEJ6WCVMZPUYLHTSQ5","short_pith_number":"pith:IJGGTCBP","schema_version":"1.0","canonical_sha256":"424c69882f1913eb0aaccbe9859e7287764e04e17f84552d3271bcc0ed29d1bb","source":{"kind":"arxiv","id":"2607.05876","version":1},"attestation_state":"computed","paper":{"title":"Think Before You Grid-Search: Floor-First Triage for LLM Serving","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.DC"],"primary_cat":"cs.PF","authors_text":"Yihua Liu","submitted_at":"2026-07-07T06:11:54Z","abstract_excerpt":"LLM serving optimization typically benchmarks many configurations and reaches for heavy profilers when latency targets are missed. We argue for the reverse discipline: estimation is the analytical layer of profiling -- without it, optimization degenerates to grid search. Floor First is a residual-driven triage workflow. Each decode step is modeled as a five-dimensional resource vector (HBM bytes, FLOPs, network bytes, network messages, KV capacity); summing within a resource and maximizing across resources gives an optimistic floor, the plain sum a pessimistic one. Where a measurement lands in"},"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":"2607.05876","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.PF","submitted_at":"2026-07-07T06:11:54Z","cross_cats_sorted":["cs.AI","cs.DC"],"title_canon_sha256":"12401ab665f63eea602b75086af0321777c2c1ef2b70d9070f880b24309474a3","abstract_canon_sha256":"0973f0b270926aeac8e2124a71bd94b03b01725849d7a5724a2e7350b90834d6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-08T01:18:49.003727Z","signature_b64":"WqLyUCDVfo5uFEHU25D0J//yVFdyDnIrwxzwQISnSHdhKy+aBwItPwk4fKJ94mQRF/5gMy2it8lWTHPq5S/xBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"424c69882f1913eb0aaccbe9859e7287764e04e17f84552d3271bcc0ed29d1bb","last_reissued_at":"2026-07-08T01:18:49.003242Z","signature_status":"signed_v1","first_computed_at":"2026-07-08T01:18:49.003242Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Think Before You Grid-Search: Floor-First Triage for LLM Serving","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.DC"],"primary_cat":"cs.PF","authors_text":"Yihua Liu","submitted_at":"2026-07-07T06:11:54Z","abstract_excerpt":"LLM serving optimization typically benchmarks many configurations and reaches for heavy profilers when latency targets are missed. We argue for the reverse discipline: estimation is the analytical layer of profiling -- without it, optimization degenerates to grid search. Floor First is a residual-driven triage workflow. Each decode step is modeled as a five-dimensional resource vector (HBM bytes, FLOPs, network bytes, network messages, KV capacity); summing within a resource and maximizing across resources gives an optimistic floor, the plain sum a pessimistic one. Where a measurement lands in"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.05876","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/2607.05876/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":"2607.05876","created_at":"2026-07-08T01:18:49.003306+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.05876v1","created_at":"2026-07-08T01:18:49.003306+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.05876","created_at":"2026-07-08T01:18:49.003306+00:00"},{"alias_kind":"pith_short_12","alias_value":"IJGGTCBPDEJ6","created_at":"2026-07-08T01:18:49.003306+00:00"},{"alias_kind":"pith_short_16","alias_value":"IJGGTCBPDEJ6WCVM","created_at":"2026-07-08T01:18:49.003306+00:00"},{"alias_kind":"pith_short_8","alias_value":"IJGGTCBP","created_at":"2026-07-08T01:18:49.003306+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/IJGGTCBPDEJ6WCVMZPUYLHTSQ5","json":"https://pith.science/pith/IJGGTCBPDEJ6WCVMZPUYLHTSQ5.json","graph_json":"https://pith.science/api/pith-number/IJGGTCBPDEJ6WCVMZPUYLHTSQ5/graph.json","events_json":"https://pith.science/api/pith-number/IJGGTCBPDEJ6WCVMZPUYLHTSQ5/events.json","paper":"https://pith.science/paper/IJGGTCBP"},"agent_actions":{"view_html":"https://pith.science/pith/IJGGTCBPDEJ6WCVMZPUYLHTSQ5","download_json":"https://pith.science/pith/IJGGTCBPDEJ6WCVMZPUYLHTSQ5.json","view_paper":"https://pith.science/paper/IJGGTCBP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.05876&json=true","fetch_graph":"https://pith.science/api/pith-number/IJGGTCBPDEJ6WCVMZPUYLHTSQ5/graph.json","fetch_events":"https://pith.science/api/pith-number/IJGGTCBPDEJ6WCVMZPUYLHTSQ5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IJGGTCBPDEJ6WCVMZPUYLHTSQ5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IJGGTCBPDEJ6WCVMZPUYLHTSQ5/action/storage_attestation","attest_author":"https://pith.science/pith/IJGGTCBPDEJ6WCVMZPUYLHTSQ5/action/author_attestation","sign_citation":"https://pith.science/pith/IJGGTCBPDEJ6WCVMZPUYLHTSQ5/action/citation_signature","submit_replication":"https://pith.science/pith/IJGGTCBPDEJ6WCVMZPUYLHTSQ5/action/replication_record"}},"created_at":"2026-07-08T01:18:49.003306+00:00","updated_at":"2026-07-08T01:18:49.003306+00:00"}