{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:T5H7IUTSM7TOTU2GCZGRJZHGFW","short_pith_number":"pith:T5H7IUTS","canonical_record":{"source":{"id":"2306.04610","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-06-07T17:22:03Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b66e787a5d71458b1f5338f4746fbbe94337d5f4aa73b53e0b3c30b1519e8833","abstract_canon_sha256":"af40c6cca65eacb80e8983a0156e3417221f543171773ba3d78c2f7097b13b78"},"schema_version":"1.0"},"canonical_sha256":"9f4ff4527267e6e9d346164d14e4e62d90214138bba1d9f4bf70603211c99387","source":{"kind":"arxiv","id":"2306.04610","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2306.04610","created_at":"2026-07-05T06:18:33Z"},{"alias_kind":"arxiv_version","alias_value":"2306.04610v1","created_at":"2026-07-05T06:18:33Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.04610","created_at":"2026-07-05T06:18:33Z"},{"alias_kind":"pith_short_12","alias_value":"T5H7IUTSM7TO","created_at":"2026-07-05T06:18:33Z"},{"alias_kind":"pith_short_16","alias_value":"T5H7IUTSM7TOTU2G","created_at":"2026-07-05T06:18:33Z"},{"alias_kind":"pith_short_8","alias_value":"T5H7IUTS","created_at":"2026-07-05T06:18:33Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:T5H7IUTSM7TOTU2GCZGRJZHGFW","target":"record","payload":{"canonical_record":{"source":{"id":"2306.04610","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-06-07T17:22:03Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b66e787a5d71458b1f5338f4746fbbe94337d5f4aa73b53e0b3c30b1519e8833","abstract_canon_sha256":"af40c6cca65eacb80e8983a0156e3417221f543171773ba3d78c2f7097b13b78"},"schema_version":"1.0"},"canonical_sha256":"9f4ff4527267e6e9d346164d14e4e62d90214138bba1d9f4bf70603211c99387","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:18:33.047589Z","signature_b64":"QvHSUAipVjH5sFDRKp0oK7l2rtSzk9zDeRrTwmNz+58JQhj0yXSHF/Rsp19kzMLYjvqmtQnC38Ekp0tmfCZ8BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9f4ff4527267e6e9d346164d14e4e62d90214138bba1d9f4bf70603211c99387","last_reissued_at":"2026-07-05T06:18:33.047173Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:18:33.047173Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2306.04610","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":"2026-07-05T06:18:33Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ODXXmQ0K4ZR6+BSXUwNqubfkCzWT0WVwfY1PIYxaVzj00RK1j/w5j5hHa0/sqY2FKcrqgUNn+4D/Nvmc7q73DA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-19T14:27:50.073380Z"},"content_sha256":"bae6be1a7ad66b2d987b2d89e9efd6b057bec32d68b048924ef131a3238ab52e","schema_version":"1.0","event_id":"sha256:bae6be1a7ad66b2d987b2d89e9efd6b057bec32d68b048924ef131a3238ab52e"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:T5H7IUTSM7TOTU2GCZGRJZHGFW","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"The Two Word Test: A Semantic Benchmark for Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Nicholas Riccardi, Rutvik H. Desai","submitted_at":"2023-06-07T17:22:03Z","abstract_excerpt":"Large Language Models (LLMs) have shown remarkable abilities recently, including passing advanced professional exams and demanding benchmark tests. This performance has led many to suggest that they are close to achieving humanlike or 'true' understanding of language, and even Artificial General Intelligence (AGI). Here, we provide a new open-source benchmark that can assess semantic abilities of LLMs using two-word phrases using a task that can be performed relatively easily by humans without advanced training. Combining multiple words into a single concept is a fundamental aspect of human la"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.04610","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/2306.04610/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":"2026-07-05T06:18:33Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ldhTNiBwbS5zoPM/nM7fjuKsC1i2f1QY9jxzHzOPJCC209TxvsFu8PXgNEJMM6ZRA89+E/AeYz7+gKDg2XSbDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-19T14:27:50.073901Z"},"content_sha256":"4016fa10d85d0ad71f3606cca138457695c00f1b71247f81ea48543f69710810","schema_version":"1.0","event_id":"sha256:4016fa10d85d0ad71f3606cca138457695c00f1b71247f81ea48543f69710810"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/T5H7IUTSM7TOTU2GCZGRJZHGFW/bundle.json","state_url":"https://pith.science/pith/T5H7IUTSM7TOTU2GCZGRJZHGFW/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/T5H7IUTSM7TOTU2GCZGRJZHGFW/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":"2026-08-19T14:27:50Z","links":{"resolver":"https://pith.science/pith/T5H7IUTSM7TOTU2GCZGRJZHGFW","bundle":"https://pith.science/pith/T5H7IUTSM7TOTU2GCZGRJZHGFW/bundle.json","state":"https://pith.science/pith/T5H7IUTSM7TOTU2GCZGRJZHGFW/state.json","well_known_bundle":"https://pith.science/.well-known/pith/T5H7IUTSM7TOTU2GCZGRJZHGFW/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:T5H7IUTSM7TOTU2GCZGRJZHGFW","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":"af40c6cca65eacb80e8983a0156e3417221f543171773ba3d78c2f7097b13b78","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-06-07T17:22:03Z","title_canon_sha256":"b66e787a5d71458b1f5338f4746fbbe94337d5f4aa73b53e0b3c30b1519e8833"},"schema_version":"1.0","source":{"id":"2306.04610","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2306.04610","created_at":"2026-07-05T06:18:33Z"},{"alias_kind":"arxiv_version","alias_value":"2306.04610v1","created_at":"2026-07-05T06:18:33Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.04610","created_at":"2026-07-05T06:18:33Z"},{"alias_kind":"pith_short_12","alias_value":"T5H7IUTSM7TO","created_at":"2026-07-05T06:18:33Z"},{"alias_kind":"pith_short_16","alias_value":"T5H7IUTSM7TOTU2G","created_at":"2026-07-05T06:18:33Z"},{"alias_kind":"pith_short_8","alias_value":"T5H7IUTS","created_at":"2026-07-05T06:18:33Z"}],"graph_snapshots":[{"event_id":"sha256:4016fa10d85d0ad71f3606cca138457695c00f1b71247f81ea48543f69710810","target":"graph","created_at":"2026-07-05T06:18:33Z","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/2306.04610/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Models (LLMs) have shown remarkable abilities recently, including passing advanced professional exams and demanding benchmark tests. This performance has led many to suggest that they are close to achieving humanlike or 'true' understanding of language, and even Artificial General Intelligence (AGI). Here, we provide a new open-source benchmark that can assess semantic abilities of LLMs using two-word phrases using a task that can be performed relatively easily by humans without advanced training. Combining multiple words into a single concept is a fundamental aspect of human la","authors_text":"Nicholas Riccardi, Rutvik H. Desai","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-06-07T17:22:03Z","title":"The Two Word Test: A Semantic Benchmark for Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.04610","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:bae6be1a7ad66b2d987b2d89e9efd6b057bec32d68b048924ef131a3238ab52e","target":"record","created_at":"2026-07-05T06:18:33Z","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":"af40c6cca65eacb80e8983a0156e3417221f543171773ba3d78c2f7097b13b78","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-06-07T17:22:03Z","title_canon_sha256":"b66e787a5d71458b1f5338f4746fbbe94337d5f4aa73b53e0b3c30b1519e8833"},"schema_version":"1.0","source":{"id":"2306.04610","kind":"arxiv","version":1}},"canonical_sha256":"9f4ff4527267e6e9d346164d14e4e62d90214138bba1d9f4bf70603211c99387","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9f4ff4527267e6e9d346164d14e4e62d90214138bba1d9f4bf70603211c99387","first_computed_at":"2026-07-05T06:18:33.047173Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:18:33.047173Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"QvHSUAipVjH5sFDRKp0oK7l2rtSzk9zDeRrTwmNz+58JQhj0yXSHF/Rsp19kzMLYjvqmtQnC38Ekp0tmfCZ8BA==","signature_status":"signed_v1","signed_at":"2026-07-05T06:18:33.047589Z","signed_message":"canonical_sha256_bytes"},"source_id":"2306.04610","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:bae6be1a7ad66b2d987b2d89e9efd6b057bec32d68b048924ef131a3238ab52e","sha256:4016fa10d85d0ad71f3606cca138457695c00f1b71247f81ea48543f69710810"],"state_sha256":"960e4d0d0b1d60c318dad8df26f48765dc0f9a75356f856b7ae9652b9ab16eda"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"x3DEEoMDR5xrShw+IvViNPHUxVAym50g6Agpuei0X73blCFP5G9mGCz+srq4hmjXavp7Cm2zF31KA5MACEu0AA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-19T14:27:50.079051Z","bundle_sha256":"33c238f084cb56f3c90546acc33319abb4e61db2cabf56baee165708ce93ba24"}}