{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:FJVTTZLLP6ICYEQK6VDMRHSTGO","short_pith_number":"pith:FJVTTZLL","canonical_record":{"source":{"id":"2109.03334","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2021-09-07T21:00:05Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"cc151a7158830161943cadd8f6d85c08d4408f101fdb443fb54d837fbf0d0830","abstract_canon_sha256":"4e44840e9350da7a7fb56b3a69454a65dbc241a25c60271b1db595776e40054a"},"schema_version":"1.0"},"canonical_sha256":"2a6b39e56b7f902c120af546c89e5333acb7d4434d1f0141265a36d396cc5718","source":{"kind":"arxiv","id":"2109.03334","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2109.03334","created_at":"2026-07-05T03:12:29Z"},{"alias_kind":"arxiv_version","alias_value":"2109.03334v1","created_at":"2026-07-05T03:12:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2109.03334","created_at":"2026-07-05T03:12:29Z"},{"alias_kind":"pith_short_12","alias_value":"FJVTTZLLP6IC","created_at":"2026-07-05T03:12:29Z"},{"alias_kind":"pith_short_16","alias_value":"FJVTTZLLP6ICYEQK","created_at":"2026-07-05T03:12:29Z"},{"alias_kind":"pith_short_8","alias_value":"FJVTTZLL","created_at":"2026-07-05T03:12:29Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:FJVTTZLLP6ICYEQK6VDMRHSTGO","target":"record","payload":{"canonical_record":{"source":{"id":"2109.03334","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2021-09-07T21:00:05Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"cc151a7158830161943cadd8f6d85c08d4408f101fdb443fb54d837fbf0d0830","abstract_canon_sha256":"4e44840e9350da7a7fb56b3a69454a65dbc241a25c60271b1db595776e40054a"},"schema_version":"1.0"},"canonical_sha256":"2a6b39e56b7f902c120af546c89e5333acb7d4434d1f0141265a36d396cc5718","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:12:29.928279Z","signature_b64":"gk2Bvj5a4ifxAtm5mDAvFQIpyGVeDPtYGaTVzfrmdFARZfxvFCGZ0M7HwIbQyXnd7yM//7Ga+wnnPigEQ+7aBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2a6b39e56b7f902c120af546c89e5333acb7d4434d1f0141265a36d396cc5718","last_reissued_at":"2026-07-05T03:12:29.927922Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:12:29.927922Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2109.03334","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":"2026-07-05T03:12:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"31N3TLu3ujbi+oujbDvTTpmQFa0kclTyF9lhdVopCrGc+OI1S8XYSTajj1BW6yEsuSVB+jI6O6+MvqQZk4d/CQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T12:20:59.926680Z"},"content_sha256":"a3d3f63a3bfc67648dff0600cf26b9abd2b8aad2d9b06dd2bd2c3ade9ab0fa4c","schema_version":"1.0","event_id":"sha256:a3d3f63a3bfc67648dff0600cf26b9abd2b8aad2d9b06dd2bd2c3ade9ab0fa4c"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:FJVTTZLLP6ICYEQK6VDMRHSTGO","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"On the Challenges of Evaluating Compositional Explanations in Multi-Hop Inference: Relevance, Completeness, and Expert Ratings","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Dan Moreno, Huitzilin Ortiz, Kelly Smith, Peter Jansen","submitted_at":"2021-09-07T21:00:05Z","abstract_excerpt":"Building compositional explanations requires models to combine two or more facts that, together, describe why the answer to a question is correct. Typically, these \"multi-hop\" explanations are evaluated relative to one (or a small number of) gold explanations. In this work, we show these evaluations substantially underestimate model performance, both in terms of the relevance of included facts, as well as the completeness of model-generated explanations, because models regularly discover and produce valid explanations that are different than gold explanations. To address this, we construct a l"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2109.03334","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/2109.03334/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":"2026-07-05T03:12:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VgqEc85udxkRgWaT+/ZaLXaoBSlH0OZV21kDXgMYxSOHeFjKdxMUbGsKeV/gpmq1w7+H6uTPaiKEXGRmWQjJCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T12:20:59.927099Z"},"content_sha256":"735be391cd6338d9eb97583af14b5866b9521c77deae72a2e7b7598c5346a0c7","schema_version":"1.0","event_id":"sha256:735be391cd6338d9eb97583af14b5866b9521c77deae72a2e7b7598c5346a0c7"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/FJVTTZLLP6ICYEQK6VDMRHSTGO/bundle.json","state_url":"https://pith.science/pith/FJVTTZLLP6ICYEQK6VDMRHSTGO/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/FJVTTZLLP6ICYEQK6VDMRHSTGO/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":"2026-07-07T12:20:59Z","links":{"resolver":"https://pith.science/pith/FJVTTZLLP6ICYEQK6VDMRHSTGO","bundle":"https://pith.science/pith/FJVTTZLLP6ICYEQK6VDMRHSTGO/bundle.json","state":"https://pith.science/pith/FJVTTZLLP6ICYEQK6VDMRHSTGO/state.json","well_known_bundle":"https://pith.science/.well-known/pith/FJVTTZLLP6ICYEQK6VDMRHSTGO/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:FJVTTZLLP6ICYEQK6VDMRHSTGO","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":"4e44840e9350da7a7fb56b3a69454a65dbc241a25c60271b1db595776e40054a","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2021-09-07T21:00:05Z","title_canon_sha256":"cc151a7158830161943cadd8f6d85c08d4408f101fdb443fb54d837fbf0d0830"},"schema_version":"1.0","source":{"id":"2109.03334","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2109.03334","created_at":"2026-07-05T03:12:29Z"},{"alias_kind":"arxiv_version","alias_value":"2109.03334v1","created_at":"2026-07-05T03:12:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2109.03334","created_at":"2026-07-05T03:12:29Z"},{"alias_kind":"pith_short_12","alias_value":"FJVTTZLLP6IC","created_at":"2026-07-05T03:12:29Z"},{"alias_kind":"pith_short_16","alias_value":"FJVTTZLLP6ICYEQK","created_at":"2026-07-05T03:12:29Z"},{"alias_kind":"pith_short_8","alias_value":"FJVTTZLL","created_at":"2026-07-05T03:12:29Z"}],"graph_snapshots":[{"event_id":"sha256:735be391cd6338d9eb97583af14b5866b9521c77deae72a2e7b7598c5346a0c7","target":"graph","created_at":"2026-07-05T03:12:29Z","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/2109.03334/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Building compositional explanations requires models to combine two or more facts that, together, describe why the answer to a question is correct. Typically, these \"multi-hop\" explanations are evaluated relative to one (or a small number of) gold explanations. In this work, we show these evaluations substantially underestimate model performance, both in terms of the relevance of included facts, as well as the completeness of model-generated explanations, because models regularly discover and produce valid explanations that are different than gold explanations. To address this, we construct a l","authors_text":"Dan Moreno, Huitzilin Ortiz, Kelly Smith, Peter Jansen","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2021-09-07T21:00:05Z","title":"On the Challenges of Evaluating Compositional Explanations in Multi-Hop Inference: Relevance, Completeness, and Expert Ratings"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2109.03334","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:a3d3f63a3bfc67648dff0600cf26b9abd2b8aad2d9b06dd2bd2c3ade9ab0fa4c","target":"record","created_at":"2026-07-05T03:12:29Z","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":"4e44840e9350da7a7fb56b3a69454a65dbc241a25c60271b1db595776e40054a","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2021-09-07T21:00:05Z","title_canon_sha256":"cc151a7158830161943cadd8f6d85c08d4408f101fdb443fb54d837fbf0d0830"},"schema_version":"1.0","source":{"id":"2109.03334","kind":"arxiv","version":1}},"canonical_sha256":"2a6b39e56b7f902c120af546c89e5333acb7d4434d1f0141265a36d396cc5718","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2a6b39e56b7f902c120af546c89e5333acb7d4434d1f0141265a36d396cc5718","first_computed_at":"2026-07-05T03:12:29.927922Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T03:12:29.927922Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"gk2Bvj5a4ifxAtm5mDAvFQIpyGVeDPtYGaTVzfrmdFARZfxvFCGZ0M7HwIbQyXnd7yM//7Ga+wnnPigEQ+7aBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T03:12:29.928279Z","signed_message":"canonical_sha256_bytes"},"source_id":"2109.03334","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a3d3f63a3bfc67648dff0600cf26b9abd2b8aad2d9b06dd2bd2c3ade9ab0fa4c","sha256:735be391cd6338d9eb97583af14b5866b9521c77deae72a2e7b7598c5346a0c7"],"state_sha256":"98e729df3ff44e45fc7dc85dcf08fa5bf4af47af587ac61405452ddc12943d75"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"V4DnYI3oQGKe2ccAcaFtH2BKJ/1NxD2YXBsjKb65FbJ4L7ogLOrvBWUR+2cDQCdfu6G6rtqDNBTRH9tDK0nDAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T12:20:59.929071Z","bundle_sha256":"b3f9e05cafdd62d409fa21a136e6b27c3480b091b3c9ef5dd961d0ef595d4343"}}