{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:S3YLKKYDK3XIGCLCPYGO4R6KOJ","short_pith_number":"pith:S3YLKKYD","canonical_record":{"source":{"id":"2102.01065","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-02-01T18:55:38Z","cross_cats_sorted":[],"title_canon_sha256":"3337a745cdc751d0183565a4cd638105d3d78deb20922b85ee0f0921c63a504f","abstract_canon_sha256":"592808cbbd2ff1d55116c8145331593a3d23b617f73b187542eb2a7ef3926ade"},"schema_version":"1.0"},"canonical_sha256":"96f0b52b0356ee8309627e0cee47ca724246fc5fd983f4b11c8f030efc0f92f2","source":{"kind":"arxiv","id":"2102.01065","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2102.01065","created_at":"2026-07-05T06:15:40Z"},{"alias_kind":"arxiv_version","alias_value":"2102.01065v3","created_at":"2026-07-05T06:15:40Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2102.01065","created_at":"2026-07-05T06:15:40Z"},{"alias_kind":"pith_short_12","alias_value":"S3YLKKYDK3XI","created_at":"2026-07-05T06:15:40Z"},{"alias_kind":"pith_short_16","alias_value":"S3YLKKYDK3XIGCLC","created_at":"2026-07-05T06:15:40Z"},{"alias_kind":"pith_short_8","alias_value":"S3YLKKYD","created_at":"2026-07-05T06:15:40Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:S3YLKKYDK3XIGCLCPYGO4R6KOJ","target":"record","payload":{"canonical_record":{"source":{"id":"2102.01065","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-02-01T18:55:38Z","cross_cats_sorted":[],"title_canon_sha256":"3337a745cdc751d0183565a4cd638105d3d78deb20922b85ee0f0921c63a504f","abstract_canon_sha256":"592808cbbd2ff1d55116c8145331593a3d23b617f73b187542eb2a7ef3926ade"},"schema_version":"1.0"},"canonical_sha256":"96f0b52b0356ee8309627e0cee47ca724246fc5fd983f4b11c8f030efc0f92f2","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:15:40.952731Z","signature_b64":"nmmLBh54xNZq0plyQEvkiAvepKUFLgfYESV0Pj1ous8xjTltDHUJ2sTPiflHfluAuOe5fETpekBuGBP7c5gWDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"96f0b52b0356ee8309627e0cee47ca724246fc5fd983f4b11c8f030efc0f92f2","last_reissued_at":"2026-07-05T06:15:40.952271Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:15:40.952271Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2102.01065","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":"2026-07-05T06:15:40Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"dYUMTETHbbZsW2EtWiG8YoM0qBj0E8xoN9b3nnBg4aW+J8g0IQ+WcMoz0X9O/id3Nrv9absw7dEnERkWjOkLAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-15T21:53:09.417951Z"},"content_sha256":"30b2557aae48ae9683b96530f9aff935c58733a5be5c4228def37e1ac1f986e0","schema_version":"1.0","event_id":"sha256:30b2557aae48ae9683b96530f9aff935c58733a5be5c4228def37e1ac1f986e0"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:S3YLKKYDK3XIGCLCPYGO4R6KOJ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Do Question Answering Modeling Improvements Hold Across Benchmarks?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Nelson F. Liu, Percy Liang, Robin Jia, Tony Lee","submitted_at":"2021-02-01T18:55:38Z","abstract_excerpt":"Do question answering (QA) modeling improvements (e.g., choice of architecture and training procedure) hold consistently across the diverse landscape of QA benchmarks? To study this question, we introduce the notion of concurrence -- two benchmarks have high concurrence on a set of modeling approaches if they rank the modeling approaches similarly. We measure the concurrence between 32 QA benchmarks on a set of 20 diverse modeling approaches and find that human-constructed benchmarks have high concurrence amongst themselves, even if their passage and question distributions are very different. "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2102.01065","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/2102.01065/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":"2026-07-05T06:15:40Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nNcOJquqn9OoS0DvoxPzoEafDiT45dJHGW3sC8rU82L4YqTZBMEq7eam6l1jZsjlyPD9zdNe3MY5waDUPw/ZCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-15T21:53:09.418520Z"},"content_sha256":"65b2c68d287e16a583930f402b532907b0032435e874589977ccd915f4ce00e1","schema_version":"1.0","event_id":"sha256:65b2c68d287e16a583930f402b532907b0032435e874589977ccd915f4ce00e1"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/S3YLKKYDK3XIGCLCPYGO4R6KOJ/bundle.json","state_url":"https://pith.science/pith/S3YLKKYDK3XIGCLCPYGO4R6KOJ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/S3YLKKYDK3XIGCLCPYGO4R6KOJ/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":"2026-08-15T21:53:09Z","links":{"resolver":"https://pith.science/pith/S3YLKKYDK3XIGCLCPYGO4R6KOJ","bundle":"https://pith.science/pith/S3YLKKYDK3XIGCLCPYGO4R6KOJ/bundle.json","state":"https://pith.science/pith/S3YLKKYDK3XIGCLCPYGO4R6KOJ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/S3YLKKYDK3XIGCLCPYGO4R6KOJ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:S3YLKKYDK3XIGCLCPYGO4R6KOJ","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":"592808cbbd2ff1d55116c8145331593a3d23b617f73b187542eb2a7ef3926ade","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-02-01T18:55:38Z","title_canon_sha256":"3337a745cdc751d0183565a4cd638105d3d78deb20922b85ee0f0921c63a504f"},"schema_version":"1.0","source":{"id":"2102.01065","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2102.01065","created_at":"2026-07-05T06:15:40Z"},{"alias_kind":"arxiv_version","alias_value":"2102.01065v3","created_at":"2026-07-05T06:15:40Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2102.01065","created_at":"2026-07-05T06:15:40Z"},{"alias_kind":"pith_short_12","alias_value":"S3YLKKYDK3XI","created_at":"2026-07-05T06:15:40Z"},{"alias_kind":"pith_short_16","alias_value":"S3YLKKYDK3XIGCLC","created_at":"2026-07-05T06:15:40Z"},{"alias_kind":"pith_short_8","alias_value":"S3YLKKYD","created_at":"2026-07-05T06:15:40Z"}],"graph_snapshots":[{"event_id":"sha256:65b2c68d287e16a583930f402b532907b0032435e874589977ccd915f4ce00e1","target":"graph","created_at":"2026-07-05T06:15:40Z","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/2102.01065/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Do question answering (QA) modeling improvements (e.g., choice of architecture and training procedure) hold consistently across the diverse landscape of QA benchmarks? To study this question, we introduce the notion of concurrence -- two benchmarks have high concurrence on a set of modeling approaches if they rank the modeling approaches similarly. We measure the concurrence between 32 QA benchmarks on a set of 20 diverse modeling approaches and find that human-constructed benchmarks have high concurrence amongst themselves, even if their passage and question distributions are very different. ","authors_text":"Nelson F. Liu, Percy Liang, Robin Jia, Tony Lee","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-02-01T18:55:38Z","title":"Do Question Answering Modeling Improvements Hold Across Benchmarks?"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2102.01065","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":"sha256:30b2557aae48ae9683b96530f9aff935c58733a5be5c4228def37e1ac1f986e0","target":"record","created_at":"2026-07-05T06:15:40Z","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":"592808cbbd2ff1d55116c8145331593a3d23b617f73b187542eb2a7ef3926ade","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-02-01T18:55:38Z","title_canon_sha256":"3337a745cdc751d0183565a4cd638105d3d78deb20922b85ee0f0921c63a504f"},"schema_version":"1.0","source":{"id":"2102.01065","kind":"arxiv","version":3}},"canonical_sha256":"96f0b52b0356ee8309627e0cee47ca724246fc5fd983f4b11c8f030efc0f92f2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"96f0b52b0356ee8309627e0cee47ca724246fc5fd983f4b11c8f030efc0f92f2","first_computed_at":"2026-07-05T06:15:40.952271Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:15:40.952271Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"nmmLBh54xNZq0plyQEvkiAvepKUFLgfYESV0Pj1ous8xjTltDHUJ2sTPiflHfluAuOe5fETpekBuGBP7c5gWDw==","signature_status":"signed_v1","signed_at":"2026-07-05T06:15:40.952731Z","signed_message":"canonical_sha256_bytes"},"source_id":"2102.01065","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:30b2557aae48ae9683b96530f9aff935c58733a5be5c4228def37e1ac1f986e0","sha256:65b2c68d287e16a583930f402b532907b0032435e874589977ccd915f4ce00e1"],"state_sha256":"ffb9564be4ce5634858e6ad7d74741239f08af168528f6e86d8ad330cb62dbd0"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Xw92WlPM1QdSti3F0ewBd6PyEJGybdFshpnLUtlfQHj27RPD9ISASo5IvyDyRhMO9XbnNfpx0P/Z8Jun6smNBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-15T21:53:09.422473Z","bundle_sha256":"04ac33178e8015aff31c73971f3a7f9d7f39bf9c231dbc835e620f7146a38630"}}