{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:5VFDH7O3EG5L5GU25YEIDBFDHN","short_pith_number":"pith:5VFDH7O3","canonical_record":{"source":{"id":"2605.24981","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-24T10:18:51Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"e782eba7e9dbd4b0f7737fcd585a4766700ef6e021e7456dcd77861bc6c102bc","abstract_canon_sha256":"dffefacbccf7f8e5072c9c0e71acd770c1322ae3b9192284d94685475ed70847"},"schema_version":"1.0"},"canonical_sha256":"ed4a33fddb21babe9a9aee088184a33b4074fd139da262ced77e5b736aea71f8","source":{"kind":"arxiv","id":"2605.24981","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.24981","created_at":"2026-05-26T01:04:09Z"},{"alias_kind":"arxiv_version","alias_value":"2605.24981v1","created_at":"2026-05-26T01:04:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.24981","created_at":"2026-05-26T01:04:09Z"},{"alias_kind":"pith_short_12","alias_value":"5VFDH7O3EG5L","created_at":"2026-05-26T01:04:09Z"},{"alias_kind":"pith_short_16","alias_value":"5VFDH7O3EG5L5GU2","created_at":"2026-05-26T01:04:09Z"},{"alias_kind":"pith_short_8","alias_value":"5VFDH7O3","created_at":"2026-05-26T01:04:09Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:5VFDH7O3EG5L5GU25YEIDBFDHN","target":"record","payload":{"canonical_record":{"source":{"id":"2605.24981","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-24T10:18:51Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"e782eba7e9dbd4b0f7737fcd585a4766700ef6e021e7456dcd77861bc6c102bc","abstract_canon_sha256":"dffefacbccf7f8e5072c9c0e71acd770c1322ae3b9192284d94685475ed70847"},"schema_version":"1.0"},"canonical_sha256":"ed4a33fddb21babe9a9aee088184a33b4074fd139da262ced77e5b736aea71f8","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-26T01:04:09.196158Z","signature_b64":"uHOTof7/sno+Z+HbJPBIw0ZLVLPDYqTUmECZ1xwfvjniMkBNnQ5rov5NSttSDD/i/QTfWWZXO7nVeKTIYOieDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ed4a33fddb21babe9a9aee088184a33b4074fd139da262ced77e5b736aea71f8","last_reissued_at":"2026-05-26T01:04:09.195484Z","signature_status":"signed_v1","first_computed_at":"2026-05-26T01:04:09.195484Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.24981","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":"2026-05-26T01:04:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Yapz2k2XgGE+3IK9y9X21S5TOyRYEslsm2hkH3jD/9z6UUo894ynUd9KOkuQu9kuOpNiWqZqh7p/do4txm/YAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-01T07:35:49.004056Z"},"content_sha256":"319206ba5b5fb4db6e03577d4004e322e780c8e89f89e353d286bf21676d5f95","schema_version":"1.0","event_id":"sha256:319206ba5b5fb4db6e03577d4004e322e780c8e89f89e353d286bf21676d5f95"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:5VFDH7O3EG5L5GU25YEIDBFDHN","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Large Language Model Selection with Limited Annotations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Andreas Kirsch, Nezihe Merve G\\\"urel, Patrik Okanovic, Torsten Hoefler, Yavuz Durmazkeser","submitted_at":"2026-05-24T10:18:51Z","abstract_excerpt":"Choosing a Large Language Model (LLM) for a given task requires comparing many strong candidates, yet standard evaluation relies on costly annotations over fixed evaluation sets. To address this challenge, we develop SELECT-LLM, the first framework for active model selection of LLMs. SELECT-LLM aims to find a small set of queries whose annotations are most informative for identifying the best LLM for a given task. To this end, we introduce a query selection rule based on expected information gain, computed from pairwise similarities between candidate model outputs. Because this rule only uses "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.24981","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/2605.24981/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":"2026-05-26T01:04:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OekSU6KZoLSVTp/MjiUtkQ0pSFuRytCMseIjpv/KctDg3jOMWrISToEyAob7k1DTUfGAhcnkEodyzqFhyQdgDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-01T07:35:49.004451Z"},"content_sha256":"4b6092245f20e5980fce92686bc7c16d1f807df6c922458f019b2036727be281","schema_version":"1.0","event_id":"sha256:4b6092245f20e5980fce92686bc7c16d1f807df6c922458f019b2036727be281"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/5VFDH7O3EG5L5GU25YEIDBFDHN/bundle.json","state_url":"https://pith.science/pith/5VFDH7O3EG5L5GU25YEIDBFDHN/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/5VFDH7O3EG5L5GU25YEIDBFDHN/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":"2026-06-01T07:35:49Z","links":{"resolver":"https://pith.science/pith/5VFDH7O3EG5L5GU25YEIDBFDHN","bundle":"https://pith.science/pith/5VFDH7O3EG5L5GU25YEIDBFDHN/bundle.json","state":"https://pith.science/pith/5VFDH7O3EG5L5GU25YEIDBFDHN/state.json","well_known_bundle":"https://pith.science/.well-known/pith/5VFDH7O3EG5L5GU25YEIDBFDHN/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:5VFDH7O3EG5L5GU25YEIDBFDHN","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":"dffefacbccf7f8e5072c9c0e71acd770c1322ae3b9192284d94685475ed70847","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-24T10:18:51Z","title_canon_sha256":"e782eba7e9dbd4b0f7737fcd585a4766700ef6e021e7456dcd77861bc6c102bc"},"schema_version":"1.0","source":{"id":"2605.24981","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.24981","created_at":"2026-05-26T01:04:09Z"},{"alias_kind":"arxiv_version","alias_value":"2605.24981v1","created_at":"2026-05-26T01:04:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.24981","created_at":"2026-05-26T01:04:09Z"},{"alias_kind":"pith_short_12","alias_value":"5VFDH7O3EG5L","created_at":"2026-05-26T01:04:09Z"},{"alias_kind":"pith_short_16","alias_value":"5VFDH7O3EG5L5GU2","created_at":"2026-05-26T01:04:09Z"},{"alias_kind":"pith_short_8","alias_value":"5VFDH7O3","created_at":"2026-05-26T01:04:09Z"}],"graph_snapshots":[{"event_id":"sha256:4b6092245f20e5980fce92686bc7c16d1f807df6c922458f019b2036727be281","target":"graph","created_at":"2026-05-26T01:04:09Z","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/2605.24981/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Choosing a Large Language Model (LLM) for a given task requires comparing many strong candidates, yet standard evaluation relies on costly annotations over fixed evaluation sets. To address this challenge, we develop SELECT-LLM, the first framework for active model selection of LLMs. SELECT-LLM aims to find a small set of queries whose annotations are most informative for identifying the best LLM for a given task. To this end, we introduce a query selection rule based on expected information gain, computed from pairwise similarities between candidate model outputs. Because this rule only uses ","authors_text":"Andreas Kirsch, Nezihe Merve G\\\"urel, Patrik Okanovic, Torsten Hoefler, Yavuz Durmazkeser","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-24T10:18:51Z","title":"Large Language Model Selection with Limited Annotations"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.24981","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":"sha256:319206ba5b5fb4db6e03577d4004e322e780c8e89f89e353d286bf21676d5f95","target":"record","created_at":"2026-05-26T01:04:09Z","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":"dffefacbccf7f8e5072c9c0e71acd770c1322ae3b9192284d94685475ed70847","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-24T10:18:51Z","title_canon_sha256":"e782eba7e9dbd4b0f7737fcd585a4766700ef6e021e7456dcd77861bc6c102bc"},"schema_version":"1.0","source":{"id":"2605.24981","kind":"arxiv","version":1}},"canonical_sha256":"ed4a33fddb21babe9a9aee088184a33b4074fd139da262ced77e5b736aea71f8","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ed4a33fddb21babe9a9aee088184a33b4074fd139da262ced77e5b736aea71f8","first_computed_at":"2026-05-26T01:04:09.195484Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-26T01:04:09.195484Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"uHOTof7/sno+Z+HbJPBIw0ZLVLPDYqTUmECZ1xwfvjniMkBNnQ5rov5NSttSDD/i/QTfWWZXO7nVeKTIYOieDA==","signature_status":"signed_v1","signed_at":"2026-05-26T01:04:09.196158Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.24981","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:319206ba5b5fb4db6e03577d4004e322e780c8e89f89e353d286bf21676d5f95","sha256:4b6092245f20e5980fce92686bc7c16d1f807df6c922458f019b2036727be281"],"state_sha256":"b1c608864b14417ed7233b754e62e1c2d3991201b6ab644f86bdb3175ffde320"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"8sP+Jssq4deAJZNtJqj0+Mx6wL7IvnYnnlyV7EXeMP9GS4G/Jt0id1EVn4hU/UuAwe/PLtg7HeJ5FVikWNipDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-01T07:35:49.006580Z","bundle_sha256":"8e95a4daebb4fd445aba47bc5010bca02d1f2de49b9039a2d5b919c4a51b2b4a"}}