{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:B5FVRDY5VXAO4JRDY5YM2JNBRF","short_pith_number":"pith:B5FVRDY5","canonical_record":{"source":{"id":"2608.03340","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-08-04T08:49:34Z","cross_cats_sorted":[],"title_canon_sha256":"5ee36bacf9eb97a273d27a15a15bda5d29d80b43593a1d38bd6c0944df1ff714","abstract_canon_sha256":"379194854d6492a7bace17cffd6479d2c16053590455185d227cb1c4e77ba7a0"},"schema_version":"1.0"},"canonical_sha256":"0f4b588f1dadc0ee2623c770cd25a189540dadcdd6fa62c306e52ad54794b7d8","source":{"kind":"arxiv","id":"2608.03340","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.03340","created_at":"2026-08-05T00:46:56Z"},{"alias_kind":"arxiv_version","alias_value":"2608.03340v1","created_at":"2026-08-05T00:46:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.03340","created_at":"2026-08-05T00:46:56Z"},{"alias_kind":"pith_short_12","alias_value":"B5FVRDY5VXAO","created_at":"2026-08-05T00:46:56Z"},{"alias_kind":"pith_short_16","alias_value":"B5FVRDY5VXAO4JRD","created_at":"2026-08-05T00:46:56Z"},{"alias_kind":"pith_short_8","alias_value":"B5FVRDY5","created_at":"2026-08-05T00:46:56Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:B5FVRDY5VXAO4JRDY5YM2JNBRF","target":"record","payload":{"canonical_record":{"source":{"id":"2608.03340","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-08-04T08:49:34Z","cross_cats_sorted":[],"title_canon_sha256":"5ee36bacf9eb97a273d27a15a15bda5d29d80b43593a1d38bd6c0944df1ff714","abstract_canon_sha256":"379194854d6492a7bace17cffd6479d2c16053590455185d227cb1c4e77ba7a0"},"schema_version":"1.0"},"canonical_sha256":"0f4b588f1dadc0ee2623c770cd25a189540dadcdd6fa62c306e52ad54794b7d8","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-05T00:46:56.588236Z","signature_b64":"kABkTCWNG4qDFMfxXMJ3lSMpvkpI+XJGuJue7YGzR8doptFvechiskQRI+MmUOrZVk1GqPdm9fEU/egpFw38CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0f4b588f1dadc0ee2623c770cd25a189540dadcdd6fa62c306e52ad54794b7d8","last_reissued_at":"2026-08-05T00:46:56.583605Z","signature_status":"signed_v1","first_computed_at":"2026-08-05T00:46:56.583605Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2608.03340","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":"2026-08-05T00:46:56Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5G7yaeIQA8gcAstmPXXNrWyHATS9fWQ1WSmLEaAkq+z0en18e+jb1x6iplVWCfw7geuO7JSm9nBjRBtYi1mUDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T11:29:33.556395Z"},"content_sha256":"6c042bcabce0b8be1e11a208212a6e4b929fb2d4de836bc3af3f8ca8ab46ce27","schema_version":"1.0","event_id":"sha256:6c042bcabce0b8be1e11a208212a6e4b929fb2d4de836bc3af3f8ca8ab46ce27"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:B5FVRDY5VXAO4JRDY5YM2JNBRF","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Benchmarking the Benchmarks: Testing the Predictive Validity of Commonsense Benchmarks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ine Gevers, Walter Daelemans","submitted_at":"2026-08-04T08:49:34Z","abstract_excerpt":"Predicting LLM's capabilities on real-world tasks is essential, yet the extent to which performance on commonsense benchmarks predicts downstream performance remains underspecified. To establish the practical usability of widely adopted commonsense benchmarks, we evaluate 23 models from six families on four established commonsense benchmarks, four reworked variants, three non-commonsense controls, and eight downstream tasks requiring implicit social, pragmatic, temporal, or physical reasoning. We compare model rankings, compute controlled correlations, and use leave-one-family-out cross-valida"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.03340","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/2608.03340/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":"2026-08-05T00:46:56Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"bUv+4wt+AOe8zof7AYNHbcLVBkVAKuVr2s/qPddL0xWnxGEw7H1gHh1Mr1WMgESIziyHmHqIuWRhEvSqUYShAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T11:29:33.556898Z"},"content_sha256":"f73c6a0dbe67e059315e63c09a6ce9caa2a0fa76a6d63c6f610b4f1b82963d5b","schema_version":"1.0","event_id":"sha256:f73c6a0dbe67e059315e63c09a6ce9caa2a0fa76a6d63c6f610b4f1b82963d5b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/B5FVRDY5VXAO4JRDY5YM2JNBRF/bundle.json","state_url":"https://pith.science/pith/B5FVRDY5VXAO4JRDY5YM2JNBRF/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/B5FVRDY5VXAO4JRDY5YM2JNBRF/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":"2026-08-10T11:29:33Z","links":{"resolver":"https://pith.science/pith/B5FVRDY5VXAO4JRDY5YM2JNBRF","bundle":"https://pith.science/pith/B5FVRDY5VXAO4JRDY5YM2JNBRF/bundle.json","state":"https://pith.science/pith/B5FVRDY5VXAO4JRDY5YM2JNBRF/state.json","well_known_bundle":"https://pith.science/.well-known/pith/B5FVRDY5VXAO4JRDY5YM2JNBRF/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:B5FVRDY5VXAO4JRDY5YM2JNBRF","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":"379194854d6492a7bace17cffd6479d2c16053590455185d227cb1c4e77ba7a0","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-08-04T08:49:34Z","title_canon_sha256":"5ee36bacf9eb97a273d27a15a15bda5d29d80b43593a1d38bd6c0944df1ff714"},"schema_version":"1.0","source":{"id":"2608.03340","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.03340","created_at":"2026-08-05T00:46:56Z"},{"alias_kind":"arxiv_version","alias_value":"2608.03340v1","created_at":"2026-08-05T00:46:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.03340","created_at":"2026-08-05T00:46:56Z"},{"alias_kind":"pith_short_12","alias_value":"B5FVRDY5VXAO","created_at":"2026-08-05T00:46:56Z"},{"alias_kind":"pith_short_16","alias_value":"B5FVRDY5VXAO4JRD","created_at":"2026-08-05T00:46:56Z"},{"alias_kind":"pith_short_8","alias_value":"B5FVRDY5","created_at":"2026-08-05T00:46:56Z"}],"graph_snapshots":[{"event_id":"sha256:f73c6a0dbe67e059315e63c09a6ce9caa2a0fa76a6d63c6f610b4f1b82963d5b","target":"graph","created_at":"2026-08-05T00:46:56Z","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/2608.03340/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Predicting LLM's capabilities on real-world tasks is essential, yet the extent to which performance on commonsense benchmarks predicts downstream performance remains underspecified. To establish the practical usability of widely adopted commonsense benchmarks, we evaluate 23 models from six families on four established commonsense benchmarks, four reworked variants, three non-commonsense controls, and eight downstream tasks requiring implicit social, pragmatic, temporal, or physical reasoning. We compare model rankings, compute controlled correlations, and use leave-one-family-out cross-valida","authors_text":"Ine Gevers, Walter Daelemans","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-08-04T08:49:34Z","title":"Benchmarking the Benchmarks: Testing the Predictive Validity of Commonsense Benchmarks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.03340","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":"sha256:6c042bcabce0b8be1e11a208212a6e4b929fb2d4de836bc3af3f8ca8ab46ce27","target":"record","created_at":"2026-08-05T00:46:56Z","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":"379194854d6492a7bace17cffd6479d2c16053590455185d227cb1c4e77ba7a0","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-08-04T08:49:34Z","title_canon_sha256":"5ee36bacf9eb97a273d27a15a15bda5d29d80b43593a1d38bd6c0944df1ff714"},"schema_version":"1.0","source":{"id":"2608.03340","kind":"arxiv","version":1}},"canonical_sha256":"0f4b588f1dadc0ee2623c770cd25a189540dadcdd6fa62c306e52ad54794b7d8","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0f4b588f1dadc0ee2623c770cd25a189540dadcdd6fa62c306e52ad54794b7d8","first_computed_at":"2026-08-05T00:46:56.583605Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-08-05T00:46:56.583605Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"kABkTCWNG4qDFMfxXMJ3lSMpvkpI+XJGuJue7YGzR8doptFvechiskQRI+MmUOrZVk1GqPdm9fEU/egpFw38CQ==","signature_status":"signed_v1","signed_at":"2026-08-05T00:46:56.588236Z","signed_message":"canonical_sha256_bytes"},"source_id":"2608.03340","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6c042bcabce0b8be1e11a208212a6e4b929fb2d4de836bc3af3f8ca8ab46ce27","sha256:f73c6a0dbe67e059315e63c09a6ce9caa2a0fa76a6d63c6f610b4f1b82963d5b"],"state_sha256":"473b0bb846808150a22b248f3c297313f614481377f46e97fd664e36e6c810e8"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"D/rb42ry6D6Sh3UxDOF5Dmc6tt09tKnRrg6K5XldnfFK4B5kMGJmW0ppEauf25fjsrYKT+s1/U7I3YjBOG1OAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-10T11:29:33.560925Z","bundle_sha256":"a5a4ad97ffc942e8f626e1ed004c6868fe7501dcbe45485a219cdde5de638480"}}