{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:TIWN2SGFVWMJJHXZKCGGZOGRYL","short_pith_number":"pith:TIWN2SGF","canonical_record":{"source":{"id":"2607.17409","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2026-07-19T21:01:34Z","cross_cats_sorted":["cs.LG","stat.ME"],"title_canon_sha256":"19e630e80abf4d474964e9d3b728ee1dce3206199374c5b63f3d0e8a8d858fc3","abstract_canon_sha256":"25f00fbed744e1b9e2055b02bf121c238e2d80e81b8914b9d1e13e7235a637e7"},"schema_version":"1.0"},"canonical_sha256":"9a2cdd48c5ad98949ef9508c6cb8d1c2e4fc3c5d52b7461b5b158f66ac071cb0","source":{"kind":"arxiv","id":"2607.17409","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.17409","created_at":"2026-07-21T01:21:31Z"},{"alias_kind":"arxiv_version","alias_value":"2607.17409v1","created_at":"2026-07-21T01:21:31Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.17409","created_at":"2026-07-21T01:21:31Z"},{"alias_kind":"pith_short_12","alias_value":"TIWN2SGFVWMJ","created_at":"2026-07-21T01:21:31Z"},{"alias_kind":"pith_short_16","alias_value":"TIWN2SGFVWMJJHXZ","created_at":"2026-07-21T01:21:31Z"},{"alias_kind":"pith_short_8","alias_value":"TIWN2SGF","created_at":"2026-07-21T01:21:31Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:TIWN2SGFVWMJJHXZKCGGZOGRYL","target":"record","payload":{"canonical_record":{"source":{"id":"2607.17409","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2026-07-19T21:01:34Z","cross_cats_sorted":["cs.LG","stat.ME"],"title_canon_sha256":"19e630e80abf4d474964e9d3b728ee1dce3206199374c5b63f3d0e8a8d858fc3","abstract_canon_sha256":"25f00fbed744e1b9e2055b02bf121c238e2d80e81b8914b9d1e13e7235a637e7"},"schema_version":"1.0"},"canonical_sha256":"9a2cdd48c5ad98949ef9508c6cb8d1c2e4fc3c5d52b7461b5b158f66ac071cb0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-21T01:21:31.530424Z","signature_b64":"KMWsbml1uTDxrGHrBedVHy9i1WDWp2LQp9dvYnKcdGBmY7i7rLVE+WB3Fsk3FTLJX0fTtkTh3sub9k2R3iLXDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9a2cdd48c5ad98949ef9508c6cb8d1c2e4fc3c5d52b7461b5b158f66ac071cb0","last_reissued_at":"2026-07-21T01:21:31.529537Z","signature_status":"signed_v1","first_computed_at":"2026-07-21T01:21:31.529537Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2607.17409","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-21T01:21:31Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gyCIAdq8MBapsgL1G9fsADISnpFG1jIvSPN42rNRjG6h//Fa1M2YA44Hm9KBiT78yOW4RkbhsGSDJ7igAYt3AQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T20:49:04.869873Z"},"content_sha256":"67cf2d259cbfb4ae0b785da624553e3cf1d09b7269c22610f8dbdb76e4662bfe","schema_version":"1.0","event_id":"sha256:67cf2d259cbfb4ae0b785da624553e3cf1d09b7269c22610f8dbdb76e4662bfe"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:TIWN2SGFVWMJJHXZKCGGZOGRYL","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Efficient Sequential Evaluation of Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","stat.ME"],"primary_cat":"stat.ML","authors_text":"Chia-Yu Hsu, Shubhanshu Shekhar","submitted_at":"2026-07-19T21:01:34Z","abstract_excerpt":"We study the problem of sequentially evaluating a new large language model (LLM) on a fixed question set using historical performance data from prior LLMs. Our goal is to construct a confidence sequence (CS) for the model's capability on this question set and to design active querying rules that shrink the CS width as quickly as possible. For CS construction, we invert a family of test supermartingales and focus on two representative approaches: a reverse information projection (RIPr)-based approach and a testing-by-betting-based approach. We first study these approaches under an oracle settin"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.17409","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/2607.17409/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-21T01:21:31Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"WblS+yB83R+0oDB0m7/MiT7uxkQ3HjcDrZ8Yvl1AXumfeno/Ylb41LWhSm8oV3NZi6xMg5k8k+XyN5NOkcezAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T20:49:04.870624Z"},"content_sha256":"06c2a935540e99edf5d77c61588d12b3e690aa0a2afdc82bda6184beeb12e25b","schema_version":"1.0","event_id":"sha256:06c2a935540e99edf5d77c61588d12b3e690aa0a2afdc82bda6184beeb12e25b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/TIWN2SGFVWMJJHXZKCGGZOGRYL/bundle.json","state_url":"https://pith.science/pith/TIWN2SGFVWMJJHXZKCGGZOGRYL/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/TIWN2SGFVWMJJHXZKCGGZOGRYL/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-04T20:49:04Z","links":{"resolver":"https://pith.science/pith/TIWN2SGFVWMJJHXZKCGGZOGRYL","bundle":"https://pith.science/pith/TIWN2SGFVWMJJHXZKCGGZOGRYL/bundle.json","state":"https://pith.science/pith/TIWN2SGFVWMJJHXZKCGGZOGRYL/state.json","well_known_bundle":"https://pith.science/.well-known/pith/TIWN2SGFVWMJJHXZKCGGZOGRYL/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:TIWN2SGFVWMJJHXZKCGGZOGRYL","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"25f00fbed744e1b9e2055b02bf121c238e2d80e81b8914b9d1e13e7235a637e7","cross_cats_sorted":["cs.LG","stat.ME"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2026-07-19T21:01:34Z","title_canon_sha256":"19e630e80abf4d474964e9d3b728ee1dce3206199374c5b63f3d0e8a8d858fc3"},"schema_version":"1.0","source":{"id":"2607.17409","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.17409","created_at":"2026-07-21T01:21:31Z"},{"alias_kind":"arxiv_version","alias_value":"2607.17409v1","created_at":"2026-07-21T01:21:31Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.17409","created_at":"2026-07-21T01:21:31Z"},{"alias_kind":"pith_short_12","alias_value":"TIWN2SGFVWMJ","created_at":"2026-07-21T01:21:31Z"},{"alias_kind":"pith_short_16","alias_value":"TIWN2SGFVWMJJHXZ","created_at":"2026-07-21T01:21:31Z"},{"alias_kind":"pith_short_8","alias_value":"TIWN2SGF","created_at":"2026-07-21T01:21:31Z"}],"graph_snapshots":[{"event_id":"sha256:06c2a935540e99edf5d77c61588d12b3e690aa0a2afdc82bda6184beeb12e25b","target":"graph","created_at":"2026-07-21T01:21:31Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2607.17409/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We study the problem of sequentially evaluating a new large language model (LLM) on a fixed question set using historical performance data from prior LLMs. Our goal is to construct a confidence sequence (CS) for the model's capability on this question set and to design active querying rules that shrink the CS width as quickly as possible. For CS construction, we invert a family of test supermartingales and focus on two representative approaches: a reverse information projection (RIPr)-based approach and a testing-by-betting-based approach. We first study these approaches under an oracle settin","authors_text":"Chia-Yu Hsu, Shubhanshu Shekhar","cross_cats":["cs.LG","stat.ME"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2026-07-19T21:01:34Z","title":"Efficient Sequential Evaluation of Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.17409","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:67cf2d259cbfb4ae0b785da624553e3cf1d09b7269c22610f8dbdb76e4662bfe","target":"record","created_at":"2026-07-21T01:21:31Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"25f00fbed744e1b9e2055b02bf121c238e2d80e81b8914b9d1e13e7235a637e7","cross_cats_sorted":["cs.LG","stat.ME"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2026-07-19T21:01:34Z","title_canon_sha256":"19e630e80abf4d474964e9d3b728ee1dce3206199374c5b63f3d0e8a8d858fc3"},"schema_version":"1.0","source":{"id":"2607.17409","kind":"arxiv","version":1}},"canonical_sha256":"9a2cdd48c5ad98949ef9508c6cb8d1c2e4fc3c5d52b7461b5b158f66ac071cb0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9a2cdd48c5ad98949ef9508c6cb8d1c2e4fc3c5d52b7461b5b158f66ac071cb0","first_computed_at":"2026-07-21T01:21:31.529537Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-21T01:21:31.529537Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"KMWsbml1uTDxrGHrBedVHy9i1WDWp2LQp9dvYnKcdGBmY7i7rLVE+WB3Fsk3FTLJX0fTtkTh3sub9k2R3iLXDg==","signature_status":"signed_v1","signed_at":"2026-07-21T01:21:31.530424Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.17409","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:67cf2d259cbfb4ae0b785da624553e3cf1d09b7269c22610f8dbdb76e4662bfe","sha256:06c2a935540e99edf5d77c61588d12b3e690aa0a2afdc82bda6184beeb12e25b"],"state_sha256":"7a8ce307c045db9774d17b009fe9ecf68ac9444ad346a1fe3e1f797c52a60ff3"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"fGzD9dOuLlO0sWO4uq+cTwc2OD5xs2rEBg+F6LYCb+Ksy0iAY6UdJ3MlX0q9Zj/thmL144KRXC0swEfhWVtJBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T20:49:04.911972Z","bundle_sha256":"166492d034497dac25f965c79f18b9d3f391fed424c9ab77c2a683f3df881e85"}}