{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:465NOZE3SMBV6O3U77MJQF3SXR","short_pith_number":"pith:465NOZE3","schema_version":"1.0","canonical_sha256":"e7bad7649b93035f3b74ffd8981772bc6aa595ec707ae0e24901740eb76da41d","source":{"kind":"arxiv","id":"2601.21522","version":2},"attestation_state":"computed","paper":{"title":"More Bang for the Buck: Improving the Inference of Large Language Models at a Fixed Budget using Reset and Discard (ReD)","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cond-mat.dis-nn","cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Barak Hirshberg, Noam Levi, Sagi Meir, Shlomi Reuveni, Tommer D. Keidar","submitted_at":"2026-01-29T10:37:32Z","abstract_excerpt":"The performance of large language models (LLMs) on verifiable tasks is usually measured by pass@k, the probability of answering a question correctly at least once in k trials. At a fixed budget, a more suitable metric is coverage@cost, the average number of unique questions answered as a function of the total number of attempts. We connect the two metrics and show that the empirically-observed power-law behavior in pass@k leads to a sublinear growth of the coverage@cost (diminishing returns). To solve this problem, we propose Reset-and-Discard (ReD), a query method of LLMs that increases cover"},"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":"2601.21522","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-01-29T10:37:32Z","cross_cats_sorted":["cond-mat.dis-nn","cs.AI","stat.ML"],"title_canon_sha256":"7d1e2890476fa610a1edb192b7afab66a548ca73360dd0f44999b11ab81c2bd7","abstract_canon_sha256":"e52f3244d5816c005958d5e8ea6095468285acf1fcc0002014153963d18f8006"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-09T01:05:13.347222Z","signature_b64":"p3SZfVf+2F7PzQuPT7ZKyxxOc7WowtETr3k8OXhOV84gdARt5T4c1SEp/lzDMyuGrIF85BCpnzDrGD2b810+Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e7bad7649b93035f3b74ffd8981772bc6aa595ec707ae0e24901740eb76da41d","last_reissued_at":"2026-06-09T01:05:13.346790Z","signature_status":"signed_v1","first_computed_at":"2026-06-09T01:05:13.346790Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"More Bang for the Buck: Improving the Inference of Large Language Models at a Fixed Budget using Reset and Discard (ReD)","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cond-mat.dis-nn","cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Barak Hirshberg, Noam Levi, Sagi Meir, Shlomi Reuveni, Tommer D. Keidar","submitted_at":"2026-01-29T10:37:32Z","abstract_excerpt":"The performance of large language models (LLMs) on verifiable tasks is usually measured by pass@k, the probability of answering a question correctly at least once in k trials. At a fixed budget, a more suitable metric is coverage@cost, the average number of unique questions answered as a function of the total number of attempts. We connect the two metrics and show that the empirically-observed power-law behavior in pass@k leads to a sublinear growth of the coverage@cost (diminishing returns). To solve this problem, we propose Reset-and-Discard (ReD), a query method of LLMs that increases cover"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2601.21522","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/2601.21522/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":"2601.21522","created_at":"2026-06-09T01:05:13.346854+00:00"},{"alias_kind":"arxiv_version","alias_value":"2601.21522v2","created_at":"2026-06-09T01:05:13.346854+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.21522","created_at":"2026-06-09T01:05:13.346854+00:00"},{"alias_kind":"pith_short_12","alias_value":"465NOZE3SMBV","created_at":"2026-06-09T01:05:13.346854+00:00"},{"alias_kind":"pith_short_16","alias_value":"465NOZE3SMBV6O3U","created_at":"2026-06-09T01:05:13.346854+00:00"},{"alias_kind":"pith_short_8","alias_value":"465NOZE3","created_at":"2026-06-09T01:05:13.346854+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/465NOZE3SMBV6O3U77MJQF3SXR","json":"https://pith.science/pith/465NOZE3SMBV6O3U77MJQF3SXR.json","graph_json":"https://pith.science/api/pith-number/465NOZE3SMBV6O3U77MJQF3SXR/graph.json","events_json":"https://pith.science/api/pith-number/465NOZE3SMBV6O3U77MJQF3SXR/events.json","paper":"https://pith.science/paper/465NOZE3"},"agent_actions":{"view_html":"https://pith.science/pith/465NOZE3SMBV6O3U77MJQF3SXR","download_json":"https://pith.science/pith/465NOZE3SMBV6O3U77MJQF3SXR.json","view_paper":"https://pith.science/paper/465NOZE3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2601.21522&json=true","fetch_graph":"https://pith.science/api/pith-number/465NOZE3SMBV6O3U77MJQF3SXR/graph.json","fetch_events":"https://pith.science/api/pith-number/465NOZE3SMBV6O3U77MJQF3SXR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/465NOZE3SMBV6O3U77MJQF3SXR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/465NOZE3SMBV6O3U77MJQF3SXR/action/storage_attestation","attest_author":"https://pith.science/pith/465NOZE3SMBV6O3U77MJQF3SXR/action/author_attestation","sign_citation":"https://pith.science/pith/465NOZE3SMBV6O3U77MJQF3SXR/action/citation_signature","submit_replication":"https://pith.science/pith/465NOZE3SMBV6O3U77MJQF3SXR/action/replication_record"}},"created_at":"2026-06-09T01:05:13.346854+00:00","updated_at":"2026-06-09T01:05:13.346854+00:00"}