{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:5US5UV4UMZTC5VKUHQLCC6CREE","short_pith_number":"pith:5US5UV4U","canonical_record":{"source":{"id":"2112.00086","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-11-30T20:36:56Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"4eb5206b09a3244f5d22e61985ad0573dd8906b1bc9e55d8fe36c90dcc3043ea","abstract_canon_sha256":"72fa9937854b7ed53232a2ade4f185f256b249879149a84d87368f40bdcd4b43"},"schema_version":"1.0"},"canonical_sha256":"ed25da579466662ed5543c162178512110d8fe027aafe05d9ac78bfe7c3d8a03","source":{"kind":"arxiv","id":"2112.00086","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2112.00086","created_at":"2026-07-05T03:36:42Z"},{"alias_kind":"arxiv_version","alias_value":"2112.00086v1","created_at":"2026-07-05T03:36:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2112.00086","created_at":"2026-07-05T03:36:42Z"},{"alias_kind":"pith_short_12","alias_value":"5US5UV4UMZTC","created_at":"2026-07-05T03:36:42Z"},{"alias_kind":"pith_short_16","alias_value":"5US5UV4UMZTC5VKU","created_at":"2026-07-05T03:36:42Z"},{"alias_kind":"pith_short_8","alias_value":"5US5UV4U","created_at":"2026-07-05T03:36:42Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:5US5UV4UMZTC5VKUHQLCC6CREE","target":"record","payload":{"canonical_record":{"source":{"id":"2112.00086","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-11-30T20:36:56Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"4eb5206b09a3244f5d22e61985ad0573dd8906b1bc9e55d8fe36c90dcc3043ea","abstract_canon_sha256":"72fa9937854b7ed53232a2ade4f185f256b249879149a84d87368f40bdcd4b43"},"schema_version":"1.0"},"canonical_sha256":"ed25da579466662ed5543c162178512110d8fe027aafe05d9ac78bfe7c3d8a03","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:36:42.367404Z","signature_b64":"pRE0hUSU8XjfjI/wPUqJ9iE6seC2FoFprDkr7vPHwG0p8HLxt3xDo8vDBpnlLf+7LSDhTOQTURQBxsfgzyIoBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ed25da579466662ed5543c162178512110d8fe027aafe05d9ac78bfe7c3d8a03","last_reissued_at":"2026-07-05T03:36:42.366984Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:36:42.366984Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2112.00086","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":"2026-07-05T03:36:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Hu/5i1MP/h9lkXqHO6YcfT9i65fmSncGqNpcbsAoWSjF9Z3gb9jt+nOW3kk4X1ZTbvmCjkRpogBu4ZYaqD1GAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T10:22:05.850273Z"},"content_sha256":"8bf9de2e3632b5837e5c7a326eebf4924c861ee76eac537470da7a761e1651a9","schema_version":"1.0","event_id":"sha256:8bf9de2e3632b5837e5c7a326eebf4924c861ee76eac537470da7a761e1651a9"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:5US5UV4UMZTC5VKUHQLCC6CREE","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Dyna-bAbI: unlocking bAbI's potential with dynamic synthetic benchmarking","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Aviad Sar-Shalom, Dafna Shahaf, Kyle Richardson, Nelson Liu, Noam Kahlon, Reut Tsarfaty, Ronen Tamari","submitted_at":"2021-11-30T20:36:56Z","abstract_excerpt":"While neural language models often perform surprisingly well on natural language understanding (NLU) tasks, their strengths and limitations remain poorly understood. Controlled synthetic tasks are thus an increasingly important resource for diagnosing model behavior. In this work we focus on story understanding, a core competency for NLU systems. However, the main synthetic resource for story understanding, the bAbI benchmark, lacks such a systematic mechanism for controllable task generation. We develop Dyna-bAbI, a dynamic framework providing fine-grained control over task generation in bAbI"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2112.00086","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/2112.00086/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":"2026-07-05T03:36:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gu79qrK5RWPJAfHBZkBqMq1g7BfasFpMZKS3pxfUtcl7tJkM9SHDZpuKBftJOmRvwJqPfgPAdtF+6kgmwOzfBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T10:22:05.851198Z"},"content_sha256":"25f841072ff18373d683e9fdc6348a9beafa3e694821f85fe87cf4e51cf96b63","schema_version":"1.0","event_id":"sha256:25f841072ff18373d683e9fdc6348a9beafa3e694821f85fe87cf4e51cf96b63"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/5US5UV4UMZTC5VKUHQLCC6CREE/bundle.json","state_url":"https://pith.science/pith/5US5UV4UMZTC5VKUHQLCC6CREE/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/5US5UV4UMZTC5VKUHQLCC6CREE/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":"2026-08-08T10:22:05Z","links":{"resolver":"https://pith.science/pith/5US5UV4UMZTC5VKUHQLCC6CREE","bundle":"https://pith.science/pith/5US5UV4UMZTC5VKUHQLCC6CREE/bundle.json","state":"https://pith.science/pith/5US5UV4UMZTC5VKUHQLCC6CREE/state.json","well_known_bundle":"https://pith.science/.well-known/pith/5US5UV4UMZTC5VKUHQLCC6CREE/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:5US5UV4UMZTC5VKUHQLCC6CREE","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":"72fa9937854b7ed53232a2ade4f185f256b249879149a84d87368f40bdcd4b43","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-11-30T20:36:56Z","title_canon_sha256":"4eb5206b09a3244f5d22e61985ad0573dd8906b1bc9e55d8fe36c90dcc3043ea"},"schema_version":"1.0","source":{"id":"2112.00086","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2112.00086","created_at":"2026-07-05T03:36:42Z"},{"alias_kind":"arxiv_version","alias_value":"2112.00086v1","created_at":"2026-07-05T03:36:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2112.00086","created_at":"2026-07-05T03:36:42Z"},{"alias_kind":"pith_short_12","alias_value":"5US5UV4UMZTC","created_at":"2026-07-05T03:36:42Z"},{"alias_kind":"pith_short_16","alias_value":"5US5UV4UMZTC5VKU","created_at":"2026-07-05T03:36:42Z"},{"alias_kind":"pith_short_8","alias_value":"5US5UV4U","created_at":"2026-07-05T03:36:42Z"}],"graph_snapshots":[{"event_id":"sha256:25f841072ff18373d683e9fdc6348a9beafa3e694821f85fe87cf4e51cf96b63","target":"graph","created_at":"2026-07-05T03:36:42Z","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/2112.00086/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"While neural language models often perform surprisingly well on natural language understanding (NLU) tasks, their strengths and limitations remain poorly understood. Controlled synthetic tasks are thus an increasingly important resource for diagnosing model behavior. In this work we focus on story understanding, a core competency for NLU systems. However, the main synthetic resource for story understanding, the bAbI benchmark, lacks such a systematic mechanism for controllable task generation. We develop Dyna-bAbI, a dynamic framework providing fine-grained control over task generation in bAbI","authors_text":"Aviad Sar-Shalom, Dafna Shahaf, Kyle Richardson, Nelson Liu, Noam Kahlon, Reut Tsarfaty, Ronen Tamari","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-11-30T20:36:56Z","title":"Dyna-bAbI: unlocking bAbI's potential with dynamic synthetic benchmarking"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2112.00086","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":"sha256:8bf9de2e3632b5837e5c7a326eebf4924c861ee76eac537470da7a761e1651a9","target":"record","created_at":"2026-07-05T03:36:42Z","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":"72fa9937854b7ed53232a2ade4f185f256b249879149a84d87368f40bdcd4b43","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-11-30T20:36:56Z","title_canon_sha256":"4eb5206b09a3244f5d22e61985ad0573dd8906b1bc9e55d8fe36c90dcc3043ea"},"schema_version":"1.0","source":{"id":"2112.00086","kind":"arxiv","version":1}},"canonical_sha256":"ed25da579466662ed5543c162178512110d8fe027aafe05d9ac78bfe7c3d8a03","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ed25da579466662ed5543c162178512110d8fe027aafe05d9ac78bfe7c3d8a03","first_computed_at":"2026-07-05T03:36:42.366984Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T03:36:42.366984Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"pRE0hUSU8XjfjI/wPUqJ9iE6seC2FoFprDkr7vPHwG0p8HLxt3xDo8vDBpnlLf+7LSDhTOQTURQBxsfgzyIoBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T03:36:42.367404Z","signed_message":"canonical_sha256_bytes"},"source_id":"2112.00086","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:8bf9de2e3632b5837e5c7a326eebf4924c861ee76eac537470da7a761e1651a9","sha256:25f841072ff18373d683e9fdc6348a9beafa3e694821f85fe87cf4e51cf96b63"],"state_sha256":"ff4a447fa19ed7b94a0cbda4b484cfef6320b3606c999051a61a74b0f8fdce4e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Z5DV6n7Msd5ST8/jhnIayRAMUGFjWQjrkax47FpVj/VOamuq4mujvIvE+rLywnI3Y+yfX/PoFYUfKFjsEdTSDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T10:22:05.857024Z","bundle_sha256":"30dffe54f945c2d2aa2d6dfe0308c3718677825dec6b838e180811b42c7bed57"}}