{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:FCT4Q56MZ5O5WZ3HYFRQSTKHGW","short_pith_number":"pith:FCT4Q56M","schema_version":"1.0","canonical_sha256":"28a7c877cccf5ddb6767c163094d4735b6be185db191cf6ef28dcf05e2b47877","source":{"kind":"arxiv","id":"2606.19719","version":1},"attestation_state":"computed","paper":{"title":"Closing the Calibration Gap in Semantic Caching","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.IR","authors_text":"Aditeya Baral, Iliya Sotirov Zhechev, Jen Agarwal, Radoslav Ralev, Srijith Rajamohan","submitted_at":"2026-06-18T02:34:33Z","abstract_excerpt":"Semantic caching cuts LLM inference costs by serving a cached response to semantically similar queries. Standard practice evaluates these systems using PR-AUC, a metric that only measures how well scores rank and ignores whether they are usable at a fixed threshold. We show this mismatch leads to systematically poor deployment choices, as models with the highest PR-AUC are often the worst in operation. We introduce Precision-Cache Hit Ratio (P-CHR) AUC, a cache-aware metric that measures precision across cache utilization levels, and Calibration Retention Rate (CRR), which captures how much of"},"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":"2606.19719","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2026-06-18T02:34:33Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"f80f75a1aedd4dddc56a845ff748055c88652e37d55d4f29aa7f7a0c503111d3","abstract_canon_sha256":"aeeb9c33218568af513efc6754c42674ddd8136e8ed86f1890f244f9840a0b07"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-19T16:12:33.225809Z","signature_b64":"PsUHe05Ml+ALMDCF4zOqXaizohyGTeOk+gANZqhWS20PRHvpdBGfNG+TzvAqpWSjajO2LpYzq3/yPHKTfhsICA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"28a7c877cccf5ddb6767c163094d4735b6be185db191cf6ef28dcf05e2b47877","last_reissued_at":"2026-06-19T16:12:33.225432Z","signature_status":"signed_v1","first_computed_at":"2026-06-19T16:12:33.225432Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Closing the Calibration Gap in Semantic Caching","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.IR","authors_text":"Aditeya Baral, Iliya Sotirov Zhechev, Jen Agarwal, Radoslav Ralev, Srijith Rajamohan","submitted_at":"2026-06-18T02:34:33Z","abstract_excerpt":"Semantic caching cuts LLM inference costs by serving a cached response to semantically similar queries. Standard practice evaluates these systems using PR-AUC, a metric that only measures how well scores rank and ignores whether they are usable at a fixed threshold. We show this mismatch leads to systematically poor deployment choices, as models with the highest PR-AUC are often the worst in operation. We introduce Precision-Cache Hit Ratio (P-CHR) AUC, a cache-aware metric that measures precision across cache utilization levels, and Calibration Retention Rate (CRR), which captures how much of"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.19719","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/2606.19719/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":"2606.19719","created_at":"2026-06-19T16:12:33.225497+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.19719v1","created_at":"2026-06-19T16:12:33.225497+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.19719","created_at":"2026-06-19T16:12:33.225497+00:00"},{"alias_kind":"pith_short_12","alias_value":"FCT4Q56MZ5O5","created_at":"2026-06-19T16:12:33.225497+00:00"},{"alias_kind":"pith_short_16","alias_value":"FCT4Q56MZ5O5WZ3H","created_at":"2026-06-19T16:12:33.225497+00:00"},{"alias_kind":"pith_short_8","alias_value":"FCT4Q56M","created_at":"2026-06-19T16:12:33.225497+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/FCT4Q56MZ5O5WZ3HYFRQSTKHGW","json":"https://pith.science/pith/FCT4Q56MZ5O5WZ3HYFRQSTKHGW.json","graph_json":"https://pith.science/api/pith-number/FCT4Q56MZ5O5WZ3HYFRQSTKHGW/graph.json","events_json":"https://pith.science/api/pith-number/FCT4Q56MZ5O5WZ3HYFRQSTKHGW/events.json","paper":"https://pith.science/paper/FCT4Q56M"},"agent_actions":{"view_html":"https://pith.science/pith/FCT4Q56MZ5O5WZ3HYFRQSTKHGW","download_json":"https://pith.science/pith/FCT4Q56MZ5O5WZ3HYFRQSTKHGW.json","view_paper":"https://pith.science/paper/FCT4Q56M","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.19719&json=true","fetch_graph":"https://pith.science/api/pith-number/FCT4Q56MZ5O5WZ3HYFRQSTKHGW/graph.json","fetch_events":"https://pith.science/api/pith-number/FCT4Q56MZ5O5WZ3HYFRQSTKHGW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FCT4Q56MZ5O5WZ3HYFRQSTKHGW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FCT4Q56MZ5O5WZ3HYFRQSTKHGW/action/storage_attestation","attest_author":"https://pith.science/pith/FCT4Q56MZ5O5WZ3HYFRQSTKHGW/action/author_attestation","sign_citation":"https://pith.science/pith/FCT4Q56MZ5O5WZ3HYFRQSTKHGW/action/citation_signature","submit_replication":"https://pith.science/pith/FCT4Q56MZ5O5WZ3HYFRQSTKHGW/action/replication_record"}},"created_at":"2026-06-19T16:12:33.225497+00:00","updated_at":"2026-06-19T16:12:33.225497+00:00"}