{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2020:IMR7D7ML7KWXO3IZVO6A33QL3B","short_pith_number":"pith:IMR7D7ML","canonical_record":{"source":{"id":"2005.00789","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-05-02T11:01:07Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"d116b1bef5502da50d88cfea4e118250950da9a76ad9db1bca053c212fb586a4","abstract_canon_sha256":"87a5575d672122b9e10c23462f105fea71a848801c9701a85f4182ae2cefe5a6"},"schema_version":"1.0"},"canonical_sha256":"4323f1fd8bfaad776d19abbc0dee0bd863a6ca55c5a5e20f7bd863d28e313686","source":{"kind":"arxiv","id":"2005.00789","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2005.00789","created_at":"2026-07-05T01:52:00Z"},{"alias_kind":"arxiv_version","alias_value":"2005.00789v3","created_at":"2026-07-05T01:52:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2005.00789","created_at":"2026-07-05T01:52:00Z"},{"alias_kind":"pith_short_12","alias_value":"IMR7D7ML7KWX","created_at":"2026-07-05T01:52:00Z"},{"alias_kind":"pith_short_16","alias_value":"IMR7D7ML7KWXO3IZ","created_at":"2026-07-05T01:52:00Z"},{"alias_kind":"pith_short_8","alias_value":"IMR7D7ML","created_at":"2026-07-05T01:52:00Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2020:IMR7D7ML7KWXO3IZVO6A33QL3B","target":"record","payload":{"canonical_record":{"source":{"id":"2005.00789","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-05-02T11:01:07Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"d116b1bef5502da50d88cfea4e118250950da9a76ad9db1bca053c212fb586a4","abstract_canon_sha256":"87a5575d672122b9e10c23462f105fea71a848801c9701a85f4182ae2cefe5a6"},"schema_version":"1.0"},"canonical_sha256":"4323f1fd8bfaad776d19abbc0dee0bd863a6ca55c5a5e20f7bd863d28e313686","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:52:00.870420Z","signature_b64":"9PytAz9QWZ6EDnt1fbfiwkp3EKA7K3+iuuBG4CuhT4KMc/+83p7gmFG+MWLgGRMAdO6SX14i+7JPMC5+rMaHAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4323f1fd8bfaad776d19abbc0dee0bd863a6ca55c5a5e20f7bd863d28e313686","last_reissued_at":"2026-07-05T01:52:00.869977Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:52:00.869977Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2005.00789","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":"2026-07-05T01:52:00Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"18p1baSKs1vU69kdJL8XCdx004q89ZV2BGNyPF03RKfKeYXmL9S9mWdKuv0PAUqqAa61Z2cT005/HYfXf+myCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T06:29:24.185283Z"},"content_sha256":"55e0f575336d6fbb922d7fe0671120d3a048d205e94ff7e1feb726ed4185164e","schema_version":"1.0","event_id":"sha256:55e0f575336d6fbb922d7fe0671120d3a048d205e94ff7e1feb726ed4185164e"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2020:IMR7D7ML7KWXO3IZVO6A33QL3B","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Is Multihop QA in DiRe Condition? Measuring and Reducing Disconnected Reasoning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Ashish Sabharwal, Harsh Trivedi, Niranjan Balasubramanian, Tushar Khot","submitted_at":"2020-05-02T11:01:07Z","abstract_excerpt":"Has there been real progress in multi-hop question-answering? Models often exploit dataset artifacts to produce correct answers, without connecting information across multiple supporting facts. This limits our ability to measure true progress and defeats the purpose of building multi-hop QA datasets. We make three contributions towards addressing this. First, we formalize such undesirable behavior as disconnected reasoning across subsets of supporting facts. This allows developing a model-agnostic probe for measuring how much any model can cheat via disconnected reasoning. Second, using a noti"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2005.00789","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/2005.00789/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":"2026-07-05T01:52:00Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"KkwTSXDAMw+HzCqfDHzGrUFCDuyex4E9w4erZfqPSnSA/Qni8Xxgt/jLZo7il6DYeuhFktzfyqSy3pa5w1LqAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T06:29:24.185849Z"},"content_sha256":"01f90d889c7865089487e53d91701bfcbc79143b82534651cd14d59b7c3555ca","schema_version":"1.0","event_id":"sha256:01f90d889c7865089487e53d91701bfcbc79143b82534651cd14d59b7c3555ca"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/IMR7D7ML7KWXO3IZVO6A33QL3B/bundle.json","state_url":"https://pith.science/pith/IMR7D7ML7KWXO3IZVO6A33QL3B/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/IMR7D7ML7KWXO3IZVO6A33QL3B/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":"2026-08-04T06:29:24Z","links":{"resolver":"https://pith.science/pith/IMR7D7ML7KWXO3IZVO6A33QL3B","bundle":"https://pith.science/pith/IMR7D7ML7KWXO3IZVO6A33QL3B/bundle.json","state":"https://pith.science/pith/IMR7D7ML7KWXO3IZVO6A33QL3B/state.json","well_known_bundle":"https://pith.science/.well-known/pith/IMR7D7ML7KWXO3IZVO6A33QL3B/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2020:IMR7D7ML7KWXO3IZVO6A33QL3B","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":"87a5575d672122b9e10c23462f105fea71a848801c9701a85f4182ae2cefe5a6","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-05-02T11:01:07Z","title_canon_sha256":"d116b1bef5502da50d88cfea4e118250950da9a76ad9db1bca053c212fb586a4"},"schema_version":"1.0","source":{"id":"2005.00789","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2005.00789","created_at":"2026-07-05T01:52:00Z"},{"alias_kind":"arxiv_version","alias_value":"2005.00789v3","created_at":"2026-07-05T01:52:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2005.00789","created_at":"2026-07-05T01:52:00Z"},{"alias_kind":"pith_short_12","alias_value":"IMR7D7ML7KWX","created_at":"2026-07-05T01:52:00Z"},{"alias_kind":"pith_short_16","alias_value":"IMR7D7ML7KWXO3IZ","created_at":"2026-07-05T01:52:00Z"},{"alias_kind":"pith_short_8","alias_value":"IMR7D7ML","created_at":"2026-07-05T01:52:00Z"}],"graph_snapshots":[{"event_id":"sha256:01f90d889c7865089487e53d91701bfcbc79143b82534651cd14d59b7c3555ca","target":"graph","created_at":"2026-07-05T01:52:00Z","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/2005.00789/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Has there been real progress in multi-hop question-answering? Models often exploit dataset artifacts to produce correct answers, without connecting information across multiple supporting facts. This limits our ability to measure true progress and defeats the purpose of building multi-hop QA datasets. We make three contributions towards addressing this. First, we formalize such undesirable behavior as disconnected reasoning across subsets of supporting facts. This allows developing a model-agnostic probe for measuring how much any model can cheat via disconnected reasoning. Second, using a noti","authors_text":"Ashish Sabharwal, Harsh Trivedi, Niranjan Balasubramanian, Tushar Khot","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-05-02T11:01:07Z","title":"Is Multihop QA in DiRe Condition? Measuring and Reducing Disconnected Reasoning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2005.00789","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":"sha256:55e0f575336d6fbb922d7fe0671120d3a048d205e94ff7e1feb726ed4185164e","target":"record","created_at":"2026-07-05T01:52:00Z","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":"87a5575d672122b9e10c23462f105fea71a848801c9701a85f4182ae2cefe5a6","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-05-02T11:01:07Z","title_canon_sha256":"d116b1bef5502da50d88cfea4e118250950da9a76ad9db1bca053c212fb586a4"},"schema_version":"1.0","source":{"id":"2005.00789","kind":"arxiv","version":3}},"canonical_sha256":"4323f1fd8bfaad776d19abbc0dee0bd863a6ca55c5a5e20f7bd863d28e313686","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4323f1fd8bfaad776d19abbc0dee0bd863a6ca55c5a5e20f7bd863d28e313686","first_computed_at":"2026-07-05T01:52:00.869977Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T01:52:00.869977Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"9PytAz9QWZ6EDnt1fbfiwkp3EKA7K3+iuuBG4CuhT4KMc/+83p7gmFG+MWLgGRMAdO6SX14i+7JPMC5+rMaHAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T01:52:00.870420Z","signed_message":"canonical_sha256_bytes"},"source_id":"2005.00789","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:55e0f575336d6fbb922d7fe0671120d3a048d205e94ff7e1feb726ed4185164e","sha256:01f90d889c7865089487e53d91701bfcbc79143b82534651cd14d59b7c3555ca"],"state_sha256":"9818b8c4f0d5e4dc1c1e8568db38b8ab996325fe7debdc864c76db45916c2b93"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"w9RGnEKttscdrocYYfqX8r97V1w38pDZcjOEZnII63TZ0bmUTltpLl+DQC5d/VqMnZ2uxD0xlXjRXqJTiG50Dw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T06:29:24.191145Z","bundle_sha256":"76fa10de7f8ad65f241dc0e185690d0e46d4b0a592e3ee5f48bb75ac393660eb"}}