{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:E6SQTFKMTBVHULLONVRMMJSBHM","short_pith_number":"pith:E6SQTFKM","schema_version":"1.0","canonical_sha256":"27a509954c986a7a2d6e6d62c626413b31b098c4362c8c5f5122fd9339058609","source":{"kind":"arxiv","id":"2310.13561","version":1},"attestation_state":"computed","paper":{"title":"Cache & Distil: Optimising API Calls to Large Language Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Alexandra Birch, Guillem Ram\\'irez, Ivan Titov, Matthias Lindemann","submitted_at":"2023-10-20T15:01:55Z","abstract_excerpt":"Large-scale deployment of generative AI tools often depends on costly API calls to a Large Language Model (LLM) to fulfil user queries. To curtail the frequency of these calls, one can employ a smaller language model -- a student -- which is continuously trained on the responses of the LLM. This student gradually gains proficiency in independently handling an increasing number of user requests, a process we term neural caching. The crucial element in neural caching is a policy that decides which requests should be processed by the student alone and which should be redirected to the LLM, subseq"},"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":"2310.13561","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-10-20T15:01:55Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"984f4beba0c048b8c8f4eec88df0eaa78139e533dceea92b6986a9ed16436b95","abstract_canon_sha256":"c99f2cdbe97f2654dfe2d1fdbb26b5dcf0791d49f438511b852adfa00f3b2ead"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:53:43.077291Z","signature_b64":"/eFCXXK/sYnEyibR2cgIzy2CSg8rWIMAUO4AZQAe3uN2qgMnYIA3YZmVQWtdYwgwtp73utUlOkUrlGIfhA1CBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"27a509954c986a7a2d6e6d62c626413b31b098c4362c8c5f5122fd9339058609","last_reissued_at":"2026-07-05T10:53:43.076801Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:53:43.076801Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Cache & Distil: Optimising API Calls to Large Language Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Alexandra Birch, Guillem Ram\\'irez, Ivan Titov, Matthias Lindemann","submitted_at":"2023-10-20T15:01:55Z","abstract_excerpt":"Large-scale deployment of generative AI tools often depends on costly API calls to a Large Language Model (LLM) to fulfil user queries. To curtail the frequency of these calls, one can employ a smaller language model -- a student -- which is continuously trained on the responses of the LLM. This student gradually gains proficiency in independently handling an increasing number of user requests, a process we term neural caching. The crucial element in neural caching is a policy that decides which requests should be processed by the student alone and which should be redirected to the LLM, subseq"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.13561","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/2310.13561/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":"2310.13561","created_at":"2026-07-05T10:53:43.076858+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.13561v1","created_at":"2026-07-05T10:53:43.076858+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.13561","created_at":"2026-07-05T10:53:43.076858+00:00"},{"alias_kind":"pith_short_12","alias_value":"E6SQTFKMTBVH","created_at":"2026-07-05T10:53:43.076858+00:00"},{"alias_kind":"pith_short_16","alias_value":"E6SQTFKMTBVHULLO","created_at":"2026-07-05T10:53:43.076858+00:00"},{"alias_kind":"pith_short_8","alias_value":"E6SQTFKM","created_at":"2026-07-05T10:53:43.076858+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.06924","citing_title":"From Sampled Outcomes to Capability Distributions: Rethinking Supervision for LLM Routing","ref_index":93,"is_internal_anchor":false},{"citing_arxiv_id":"2502.18036","citing_title":"Harnessing Multiple Large Language Models: A Survey on LLM Ensemble","ref_index":40,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/E6SQTFKMTBVHULLONVRMMJSBHM","json":"https://pith.science/pith/E6SQTFKMTBVHULLONVRMMJSBHM.json","graph_json":"https://pith.science/api/pith-number/E6SQTFKMTBVHULLONVRMMJSBHM/graph.json","events_json":"https://pith.science/api/pith-number/E6SQTFKMTBVHULLONVRMMJSBHM/events.json","paper":"https://pith.science/paper/E6SQTFKM"},"agent_actions":{"view_html":"https://pith.science/pith/E6SQTFKMTBVHULLONVRMMJSBHM","download_json":"https://pith.science/pith/E6SQTFKMTBVHULLONVRMMJSBHM.json","view_paper":"https://pith.science/paper/E6SQTFKM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.13561&json=true","fetch_graph":"https://pith.science/api/pith-number/E6SQTFKMTBVHULLONVRMMJSBHM/graph.json","fetch_events":"https://pith.science/api/pith-number/E6SQTFKMTBVHULLONVRMMJSBHM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/E6SQTFKMTBVHULLONVRMMJSBHM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/E6SQTFKMTBVHULLONVRMMJSBHM/action/storage_attestation","attest_author":"https://pith.science/pith/E6SQTFKMTBVHULLONVRMMJSBHM/action/author_attestation","sign_citation":"https://pith.science/pith/E6SQTFKMTBVHULLONVRMMJSBHM/action/citation_signature","submit_replication":"https://pith.science/pith/E6SQTFKMTBVHULLONVRMMJSBHM/action/replication_record"}},"created_at":"2026-07-05T10:53:43.076858+00:00","updated_at":"2026-07-05T10:53:43.076858+00:00"}