{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:J77RV6BVJXFMBCPIKQ4E6JN32X","short_pith_number":"pith:J77RV6BV","canonical_record":{"source":{"id":"2108.06130","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-08-13T09:12:27Z","cross_cats_sorted":["cs.IR"],"title_canon_sha256":"9176fb02d5d12ce0a75c15e0831de81af75210692f3fe2bddd83aea7840797ae","abstract_canon_sha256":"ffa4f27db2ab9587755f9f71348c4363e289c036f688c327fd7ffd9b94e2b87c"},"schema_version":"1.0"},"canonical_sha256":"4fff1af8354dcac089e854384f25bbd5c21059d05248ba338c9f03bfe6a47f95","source":{"kind":"arxiv","id":"2108.06130","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2108.06130","created_at":"2026-07-05T03:24:31Z"},{"alias_kind":"arxiv_version","alias_value":"2108.06130v3","created_at":"2026-07-05T03:24:31Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2108.06130","created_at":"2026-07-05T03:24:31Z"},{"alias_kind":"pith_short_12","alias_value":"J77RV6BVJXFM","created_at":"2026-07-05T03:24:31Z"},{"alias_kind":"pith_short_16","alias_value":"J77RV6BVJXFMBCPI","created_at":"2026-07-05T03:24:31Z"},{"alias_kind":"pith_short_8","alias_value":"J77RV6BV","created_at":"2026-07-05T03:24:31Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:J77RV6BVJXFMBCPIKQ4E6JN32X","target":"record","payload":{"canonical_record":{"source":{"id":"2108.06130","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-08-13T09:12:27Z","cross_cats_sorted":["cs.IR"],"title_canon_sha256":"9176fb02d5d12ce0a75c15e0831de81af75210692f3fe2bddd83aea7840797ae","abstract_canon_sha256":"ffa4f27db2ab9587755f9f71348c4363e289c036f688c327fd7ffd9b94e2b87c"},"schema_version":"1.0"},"canonical_sha256":"4fff1af8354dcac089e854384f25bbd5c21059d05248ba338c9f03bfe6a47f95","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:24:31.078212Z","signature_b64":"fT5KxeSfRhi01cIukc2hri332KEhG08xL85Fa+kOeYDy0BaTWnMdL9Gyi3Hp6M2jqN1yKKd27qB5zgUMkbXrCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4fff1af8354dcac089e854384f25bbd5c21059d05248ba338c9f03bfe6a47f95","last_reissued_at":"2026-07-05T03:24:31.077696Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:24:31.077696Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2108.06130","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":"2026-07-05T03:24:31Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"MKPXWX1LYrYOyjgEjyyhqvsJdfGvYyG20yKb8YuYlH+gVYlxpKkhb+qLCOcNdXC6tRuDqskSdskcLypUIRrQCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-26T16:03:43.581365Z"},"content_sha256":"bca145d752987901f1919e6aecd7eb5bcf37d348a2f7494334ea767248ec7617","schema_version":"1.0","event_id":"sha256:bca145d752987901f1919e6aecd7eb5bcf37d348a2f7494334ea767248ec7617"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:J77RV6BVJXFMBCPIKQ4E6JN32X","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Semantic Answer Similarity for Evaluating Question Answering Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.IR"],"primary_cat":"cs.CL","authors_text":"Julian Gutsch, Julian Risch, Malte Pietsch, Timo M\\\"oller","submitted_at":"2021-08-13T09:12:27Z","abstract_excerpt":"The evaluation of question answering models compares ground-truth annotations with model predictions. However, as of today, this comparison is mostly lexical-based and therefore misses out on answers that have no lexical overlap but are still semantically similar, thus treating correct answers as false. This underestimation of the true performance of models hinders user acceptance in applications and complicates a fair comparison of different models. Therefore, there is a need for an evaluation metric that is based on semantics instead of pure string similarity. In this short paper, we present"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2108.06130","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/2108.06130/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":"2026-07-05T03:24:31Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6WYqX4n5ibo+UTzkCs62U19iq/fTHKzWyELJlP26U+U8XB30Et5zaPQHlcJpgPSNeAqIgskZBUyeEtanRQ2/Aw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-26T16:03:43.581747Z"},"content_sha256":"d04390643c9d6c21d90f13896d4cfafe03e2cec54ec09e31d00ea7276587f9fa","schema_version":"1.0","event_id":"sha256:d04390643c9d6c21d90f13896d4cfafe03e2cec54ec09e31d00ea7276587f9fa"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/J77RV6BVJXFMBCPIKQ4E6JN32X/bundle.json","state_url":"https://pith.science/pith/J77RV6BVJXFMBCPIKQ4E6JN32X/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/J77RV6BVJXFMBCPIKQ4E6JN32X/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":"2026-07-26T16:03:43Z","links":{"resolver":"https://pith.science/pith/J77RV6BVJXFMBCPIKQ4E6JN32X","bundle":"https://pith.science/pith/J77RV6BVJXFMBCPIKQ4E6JN32X/bundle.json","state":"https://pith.science/pith/J77RV6BVJXFMBCPIKQ4E6JN32X/state.json","well_known_bundle":"https://pith.science/.well-known/pith/J77RV6BVJXFMBCPIKQ4E6JN32X/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:J77RV6BVJXFMBCPIKQ4E6JN32X","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":"ffa4f27db2ab9587755f9f71348c4363e289c036f688c327fd7ffd9b94e2b87c","cross_cats_sorted":["cs.IR"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-08-13T09:12:27Z","title_canon_sha256":"9176fb02d5d12ce0a75c15e0831de81af75210692f3fe2bddd83aea7840797ae"},"schema_version":"1.0","source":{"id":"2108.06130","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2108.06130","created_at":"2026-07-05T03:24:31Z"},{"alias_kind":"arxiv_version","alias_value":"2108.06130v3","created_at":"2026-07-05T03:24:31Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2108.06130","created_at":"2026-07-05T03:24:31Z"},{"alias_kind":"pith_short_12","alias_value":"J77RV6BVJXFM","created_at":"2026-07-05T03:24:31Z"},{"alias_kind":"pith_short_16","alias_value":"J77RV6BVJXFMBCPI","created_at":"2026-07-05T03:24:31Z"},{"alias_kind":"pith_short_8","alias_value":"J77RV6BV","created_at":"2026-07-05T03:24:31Z"}],"graph_snapshots":[{"event_id":"sha256:d04390643c9d6c21d90f13896d4cfafe03e2cec54ec09e31d00ea7276587f9fa","target":"graph","created_at":"2026-07-05T03:24:31Z","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/2108.06130/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The evaluation of question answering models compares ground-truth annotations with model predictions. However, as of today, this comparison is mostly lexical-based and therefore misses out on answers that have no lexical overlap but are still semantically similar, thus treating correct answers as false. This underestimation of the true performance of models hinders user acceptance in applications and complicates a fair comparison of different models. Therefore, there is a need for an evaluation metric that is based on semantics instead of pure string similarity. In this short paper, we present","authors_text":"Julian Gutsch, Julian Risch, Malte Pietsch, Timo M\\\"oller","cross_cats":["cs.IR"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-08-13T09:12:27Z","title":"Semantic Answer Similarity for Evaluating Question Answering Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2108.06130","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:bca145d752987901f1919e6aecd7eb5bcf37d348a2f7494334ea767248ec7617","target":"record","created_at":"2026-07-05T03:24:31Z","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":"ffa4f27db2ab9587755f9f71348c4363e289c036f688c327fd7ffd9b94e2b87c","cross_cats_sorted":["cs.IR"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-08-13T09:12:27Z","title_canon_sha256":"9176fb02d5d12ce0a75c15e0831de81af75210692f3fe2bddd83aea7840797ae"},"schema_version":"1.0","source":{"id":"2108.06130","kind":"arxiv","version":3}},"canonical_sha256":"4fff1af8354dcac089e854384f25bbd5c21059d05248ba338c9f03bfe6a47f95","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4fff1af8354dcac089e854384f25bbd5c21059d05248ba338c9f03bfe6a47f95","first_computed_at":"2026-07-05T03:24:31.077696Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T03:24:31.077696Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"fT5KxeSfRhi01cIukc2hri332KEhG08xL85Fa+kOeYDy0BaTWnMdL9Gyi3Hp6M2jqN1yKKd27qB5zgUMkbXrCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T03:24:31.078212Z","signed_message":"canonical_sha256_bytes"},"source_id":"2108.06130","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:bca145d752987901f1919e6aecd7eb5bcf37d348a2f7494334ea767248ec7617","sha256:d04390643c9d6c21d90f13896d4cfafe03e2cec54ec09e31d00ea7276587f9fa"],"state_sha256":"356bbebbb27f6a10322f95f629b0b62fd2067b7cf81f7364c38f04ebd8c55b89"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"L1mirt27YJKzWmr5CsulRzCWIOBqTPi/B6niOWkBcQZ5JUMtyCIFwGbfKUZh+G4Uz+FWToDxnPA73myv2wFfCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-26T16:03:43.584299Z","bundle_sha256":"a8b105c8cc09c6f7901a7e5de7deabf812609bec744017870fb30de81dd447ff"}}