{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:IQZRABNHMOCYID7FYL5Q3GSOJT","short_pith_number":"pith:IQZRABNH","canonical_record":{"source":{"id":"2511.04689","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-10-26T03:54:12Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"34ceadfb5663048b1607c9ae8d92d5b3079648338abca90558c37d22a5c62142","abstract_canon_sha256":"21fa915fddb4b83ca8d81fb0893a080040166d87ec626e90f35b4519b198793a"},"schema_version":"1.0"},"canonical_sha256":"44331005a76385840fe5c2fb0d9a4e4cda605d7832e72ba19934d140f123041a","source":{"kind":"arxiv","id":"2511.04689","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2511.04689","created_at":"2026-07-15T00:21:13Z"},{"alias_kind":"arxiv_version","alias_value":"2511.04689v3","created_at":"2026-07-15T00:21:13Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2511.04689","created_at":"2026-07-15T00:21:13Z"},{"alias_kind":"pith_short_12","alias_value":"IQZRABNHMOCY","created_at":"2026-07-15T00:21:13Z"},{"alias_kind":"pith_short_16","alias_value":"IQZRABNHMOCYID7F","created_at":"2026-07-15T00:21:13Z"},{"alias_kind":"pith_short_8","alias_value":"IQZRABNH","created_at":"2026-07-15T00:21:13Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:IQZRABNHMOCYID7FYL5Q3GSOJT","target":"record","payload":{"canonical_record":{"source":{"id":"2511.04689","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-10-26T03:54:12Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"34ceadfb5663048b1607c9ae8d92d5b3079648338abca90558c37d22a5c62142","abstract_canon_sha256":"21fa915fddb4b83ca8d81fb0893a080040166d87ec626e90f35b4519b198793a"},"schema_version":"1.0"},"canonical_sha256":"44331005a76385840fe5c2fb0d9a4e4cda605d7832e72ba19934d140f123041a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-15T00:21:13.643980Z","signature_b64":"8tD2qqnmhQpJO43zb7+lHRAEYIyODBXTfy8Ddd5Cz9/DaGmfeyCkoPD4ykQTsrhlxTchBh3q46lSsCYGmDWyCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"44331005a76385840fe5c2fb0d9a4e4cda605d7832e72ba19934d140f123041a","last_reissued_at":"2026-07-15T00:21:13.643010Z","signature_status":"signed_v1","first_computed_at":"2026-07-15T00:21:13.643010Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2511.04689","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-15T00:21:13Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xYQcHzImkqZmtotzgxKyxbHI8odoQ4mLNkqB/MtDTj9TJf+s7CqSMxQVNccnUJSicaUN5jm9E+zvkbZW5Q5VBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-31T20:01:30.527252Z"},"content_sha256":"f96d99e56c901c66f5a98ccd455fede824607ebb53a87b7354631a8e77ef7fb1","schema_version":"1.0","event_id":"sha256:f96d99e56c901c66f5a98ccd455fede824607ebb53a87b7354631a8e77ef7fb1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:IQZRABNHMOCYID7FYL5Q3GSOJT","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Adaptive Testing for LLM Evaluation: A Psychometric Alternative to Static Benchmarks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Nitesh V. Chawla, Peiyu Li, Ronald Metoyer, Si Chen, Ting Hua, Xiuxiu Tang, Ying Cheng","submitted_at":"2025-10-26T03:54:12Z","abstract_excerpt":"Evaluating large language models (LLMs) typically requires thousands of benchmark items, making the process expensive, slow, and increasingly impractical at scale. Existing evaluation protocols rely on average accuracy over fixed item sets, treating all items as equally informative despite substantial variation in difficulty and discrimination. We introduce ATLAS, an adaptive testing framework based on Item Response Theory (IRT) that estimates model ability using Fisher information-guided item selection. ATLAS reduces the number of required items by up to 90% while maintaining measurement prec"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2511.04689","kind":"arxiv","version":3},"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/2511.04689/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":"2026-07-15T00:21:13Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"e4g/IiIqAcBTRqqN49N4FcCaWhV4fEDRPuSZVBG1xe8vbXp8GpCklzL+/J50mfiIew1f06Tx69AgSYbcc3HgCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-31T20:01:30.527739Z"},"content_sha256":"91d1cbc08ec1be560f7a29737b03ec4eb7a0d85cdfbafba6a16fd7c341f44b52","schema_version":"1.0","event_id":"sha256:91d1cbc08ec1be560f7a29737b03ec4eb7a0d85cdfbafba6a16fd7c341f44b52"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/IQZRABNHMOCYID7FYL5Q3GSOJT/bundle.json","state_url":"https://pith.science/pith/IQZRABNHMOCYID7FYL5Q3GSOJT/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/IQZRABNHMOCYID7FYL5Q3GSOJT/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":"2026-07-31T20:01:30Z","links":{"resolver":"https://pith.science/pith/IQZRABNHMOCYID7FYL5Q3GSOJT","bundle":"https://pith.science/pith/IQZRABNHMOCYID7FYL5Q3GSOJT/bundle.json","state":"https://pith.science/pith/IQZRABNHMOCYID7FYL5Q3GSOJT/state.json","well_known_bundle":"https://pith.science/.well-known/pith/IQZRABNHMOCYID7FYL5Q3GSOJT/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:IQZRABNHMOCYID7FYL5Q3GSOJT","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":"21fa915fddb4b83ca8d81fb0893a080040166d87ec626e90f35b4519b198793a","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-10-26T03:54:12Z","title_canon_sha256":"34ceadfb5663048b1607c9ae8d92d5b3079648338abca90558c37d22a5c62142"},"schema_version":"1.0","source":{"id":"2511.04689","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2511.04689","created_at":"2026-07-15T00:21:13Z"},{"alias_kind":"arxiv_version","alias_value":"2511.04689v3","created_at":"2026-07-15T00:21:13Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2511.04689","created_at":"2026-07-15T00:21:13Z"},{"alias_kind":"pith_short_12","alias_value":"IQZRABNHMOCY","created_at":"2026-07-15T00:21:13Z"},{"alias_kind":"pith_short_16","alias_value":"IQZRABNHMOCYID7F","created_at":"2026-07-15T00:21:13Z"},{"alias_kind":"pith_short_8","alias_value":"IQZRABNH","created_at":"2026-07-15T00:21:13Z"}],"graph_snapshots":[{"event_id":"sha256:91d1cbc08ec1be560f7a29737b03ec4eb7a0d85cdfbafba6a16fd7c341f44b52","target":"graph","created_at":"2026-07-15T00:21:13Z","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/2511.04689/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Evaluating large language models (LLMs) typically requires thousands of benchmark items, making the process expensive, slow, and increasingly impractical at scale. Existing evaluation protocols rely on average accuracy over fixed item sets, treating all items as equally informative despite substantial variation in difficulty and discrimination. We introduce ATLAS, an adaptive testing framework based on Item Response Theory (IRT) that estimates model ability using Fisher information-guided item selection. ATLAS reduces the number of required items by up to 90% while maintaining measurement prec","authors_text":"Nitesh V. Chawla, Peiyu Li, Ronald Metoyer, Si Chen, Ting Hua, Xiuxiu Tang, Ying Cheng","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-10-26T03:54:12Z","title":"Adaptive Testing for LLM Evaluation: A Psychometric Alternative to Static Benchmarks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2511.04689","kind":"arxiv","version":3},"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:f96d99e56c901c66f5a98ccd455fede824607ebb53a87b7354631a8e77ef7fb1","target":"record","created_at":"2026-07-15T00:21:13Z","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":"21fa915fddb4b83ca8d81fb0893a080040166d87ec626e90f35b4519b198793a","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-10-26T03:54:12Z","title_canon_sha256":"34ceadfb5663048b1607c9ae8d92d5b3079648338abca90558c37d22a5c62142"},"schema_version":"1.0","source":{"id":"2511.04689","kind":"arxiv","version":3}},"canonical_sha256":"44331005a76385840fe5c2fb0d9a4e4cda605d7832e72ba19934d140f123041a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"44331005a76385840fe5c2fb0d9a4e4cda605d7832e72ba19934d140f123041a","first_computed_at":"2026-07-15T00:21:13.643010Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-15T00:21:13.643010Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"8tD2qqnmhQpJO43zb7+lHRAEYIyODBXTfy8Ddd5Cz9/DaGmfeyCkoPD4ykQTsrhlxTchBh3q46lSsCYGmDWyCQ==","signature_status":"signed_v1","signed_at":"2026-07-15T00:21:13.643980Z","signed_message":"canonical_sha256_bytes"},"source_id":"2511.04689","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f96d99e56c901c66f5a98ccd455fede824607ebb53a87b7354631a8e77ef7fb1","sha256:91d1cbc08ec1be560f7a29737b03ec4eb7a0d85cdfbafba6a16fd7c341f44b52"],"state_sha256":"80579f8227199bce3e8c22994469e124ea8e3bcc871619385f6abd6ea1ccff7a"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"moUy6FD4v6559XhQ7y19q8T51tIVtwTCkyimUm+uCFIggrUdqQzp4vEvkLSgIeTSv9v3LYKsC5/Yxv6OcVbJBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-31T20:01:30.531474Z","bundle_sha256":"a9538054ce9745efa3080e096cce0e4be9e086e2823c377ba6a66a71f1b24430"}}