{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:YK4EUGHCAOCJT22EEAGXFCCC7E","short_pith_number":"pith:YK4EUGHC","canonical_record":{"source":{"id":"2307.02477","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-07-05T17:50:42Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"80dd73d9a5eccbbca04c81914a1cca0fda1e90ef6e6a7323e885911ae610d9c4","abstract_canon_sha256":"cd8974d3ab18e555109b0d06d651c279d3835606f6e2bd8d9b9f15922626a77c"},"schema_version":"1.0"},"canonical_sha256":"c2b84a18e2038499eb44200d728842f9342c300c0ae043e40de562cf603578ed","source":{"kind":"arxiv","id":"2307.02477","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2307.02477","created_at":"2026-07-05T08:02:00Z"},{"alias_kind":"arxiv_version","alias_value":"2307.02477v3","created_at":"2026-07-05T08:02:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2307.02477","created_at":"2026-07-05T08:02:00Z"},{"alias_kind":"pith_short_12","alias_value":"YK4EUGHCAOCJ","created_at":"2026-07-05T08:02:00Z"},{"alias_kind":"pith_short_16","alias_value":"YK4EUGHCAOCJT22E","created_at":"2026-07-05T08:02:00Z"},{"alias_kind":"pith_short_8","alias_value":"YK4EUGHC","created_at":"2026-07-05T08:02:00Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:YK4EUGHCAOCJT22EEAGXFCCC7E","target":"record","payload":{"canonical_record":{"source":{"id":"2307.02477","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-07-05T17:50:42Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"80dd73d9a5eccbbca04c81914a1cca0fda1e90ef6e6a7323e885911ae610d9c4","abstract_canon_sha256":"cd8974d3ab18e555109b0d06d651c279d3835606f6e2bd8d9b9f15922626a77c"},"schema_version":"1.0"},"canonical_sha256":"c2b84a18e2038499eb44200d728842f9342c300c0ae043e40de562cf603578ed","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:02:00.022934Z","signature_b64":"NZ0eG3mxBGIAoaSetNEaML0wlOp1ie8JDIDafAGv21fam350mWjdDqENZoYXPw3e+nf0YxH0JJacawE5g6hvAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c2b84a18e2038499eb44200d728842f9342c300c0ae043e40de562cf603578ed","last_reissued_at":"2026-07-05T08:02:00.022396Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:02:00.022396Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2307.02477","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":"2026-07-05T08:02:00Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"KspiA6vUrHcJ5ex8iJIE3o6m1m8jtHERPSFEvVPQmko1xfODemtXimUlnz3ZanYMJDs6mLWVQD9B4lgnw0a+Cw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T08:34:03.463692Z"},"content_sha256":"003cd02f1a28db2263d21d3a7e723852fb63a09d13e4365f124c626c0b823e8e","schema_version":"1.0","event_id":"sha256:003cd02f1a28db2263d21d3a7e723852fb63a09d13e4365f124c626c0b823e8e"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:YK4EUGHCAOCJT22EEAGXFCCC7E","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Reasoning or Reciting? Exploring the Capabilities and Limitations of Language Models Through Counterfactual Tasks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Alexis Ross, Bailin Wang, Boyuan Chen, Ekin Aky\\\"urek, Jacob Andreas, Linlu Qiu, Najoung Kim, Yoon Kim, Zhaofeng Wu","submitted_at":"2023-07-05T17:50:42Z","abstract_excerpt":"The impressive performance of recent language models across a wide range of tasks suggests that they possess a degree of abstract reasoning skills. Are these skills general and transferable, or specialized to specific tasks seen during pretraining? To disentangle these effects, we propose an evaluation framework based on \"counterfactual\" task variants that deviate from the default assumptions underlying standard tasks. Across a suite of 11 tasks, we observe nontrivial performance on the counterfactual variants, but nevertheless find that performance substantially and consistently degrades comp"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2307.02477","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/2307.02477/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":"2026-07-05T08:02:00Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"AX5Op+Dm8mViVQWbc4jdpwWeH8ZZ56EAtO4DIRwLmQ914onWkb1SlIRXUORSyg5vfhRuetf6/UeRyupCSB4ZCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T08:34:03.464234Z"},"content_sha256":"0890ce24c511024a1d8ccac24c4a88d3351803de45154cb8aadf2ed443092e6a","schema_version":"1.0","event_id":"sha256:0890ce24c511024a1d8ccac24c4a88d3351803de45154cb8aadf2ed443092e6a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/YK4EUGHCAOCJT22EEAGXFCCC7E/bundle.json","state_url":"https://pith.science/pith/YK4EUGHCAOCJT22EEAGXFCCC7E/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/YK4EUGHCAOCJT22EEAGXFCCC7E/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":"2026-08-08T08:34:03Z","links":{"resolver":"https://pith.science/pith/YK4EUGHCAOCJT22EEAGXFCCC7E","bundle":"https://pith.science/pith/YK4EUGHCAOCJT22EEAGXFCCC7E/bundle.json","state":"https://pith.science/pith/YK4EUGHCAOCJT22EEAGXFCCC7E/state.json","well_known_bundle":"https://pith.science/.well-known/pith/YK4EUGHCAOCJT22EEAGXFCCC7E/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:YK4EUGHCAOCJT22EEAGXFCCC7E","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":"cd8974d3ab18e555109b0d06d651c279d3835606f6e2bd8d9b9f15922626a77c","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-07-05T17:50:42Z","title_canon_sha256":"80dd73d9a5eccbbca04c81914a1cca0fda1e90ef6e6a7323e885911ae610d9c4"},"schema_version":"1.0","source":{"id":"2307.02477","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2307.02477","created_at":"2026-07-05T08:02:00Z"},{"alias_kind":"arxiv_version","alias_value":"2307.02477v3","created_at":"2026-07-05T08:02:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2307.02477","created_at":"2026-07-05T08:02:00Z"},{"alias_kind":"pith_short_12","alias_value":"YK4EUGHCAOCJ","created_at":"2026-07-05T08:02:00Z"},{"alias_kind":"pith_short_16","alias_value":"YK4EUGHCAOCJT22E","created_at":"2026-07-05T08:02:00Z"},{"alias_kind":"pith_short_8","alias_value":"YK4EUGHC","created_at":"2026-07-05T08:02:00Z"}],"graph_snapshots":[{"event_id":"sha256:0890ce24c511024a1d8ccac24c4a88d3351803de45154cb8aadf2ed443092e6a","target":"graph","created_at":"2026-07-05T08:02:00Z","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/2307.02477/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The impressive performance of recent language models across a wide range of tasks suggests that they possess a degree of abstract reasoning skills. Are these skills general and transferable, or specialized to specific tasks seen during pretraining? To disentangle these effects, we propose an evaluation framework based on \"counterfactual\" task variants that deviate from the default assumptions underlying standard tasks. Across a suite of 11 tasks, we observe nontrivial performance on the counterfactual variants, but nevertheless find that performance substantially and consistently degrades comp","authors_text":"Alexis Ross, Bailin Wang, Boyuan Chen, Ekin Aky\\\"urek, Jacob Andreas, Linlu Qiu, Najoung Kim, Yoon Kim, Zhaofeng Wu","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-07-05T17:50:42Z","title":"Reasoning or Reciting? Exploring the Capabilities and Limitations of Language Models Through Counterfactual Tasks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2307.02477","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":"sha256:003cd02f1a28db2263d21d3a7e723852fb63a09d13e4365f124c626c0b823e8e","target":"record","created_at":"2026-07-05T08:02:00Z","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":"cd8974d3ab18e555109b0d06d651c279d3835606f6e2bd8d9b9f15922626a77c","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-07-05T17:50:42Z","title_canon_sha256":"80dd73d9a5eccbbca04c81914a1cca0fda1e90ef6e6a7323e885911ae610d9c4"},"schema_version":"1.0","source":{"id":"2307.02477","kind":"arxiv","version":3}},"canonical_sha256":"c2b84a18e2038499eb44200d728842f9342c300c0ae043e40de562cf603578ed","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c2b84a18e2038499eb44200d728842f9342c300c0ae043e40de562cf603578ed","first_computed_at":"2026-07-05T08:02:00.022396Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:02:00.022396Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"NZ0eG3mxBGIAoaSetNEaML0wlOp1ie8JDIDafAGv21fam350mWjdDqENZoYXPw3e+nf0YxH0JJacawE5g6hvAA==","signature_status":"signed_v1","signed_at":"2026-07-05T08:02:00.022934Z","signed_message":"canonical_sha256_bytes"},"source_id":"2307.02477","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:003cd02f1a28db2263d21d3a7e723852fb63a09d13e4365f124c626c0b823e8e","sha256:0890ce24c511024a1d8ccac24c4a88d3351803de45154cb8aadf2ed443092e6a"],"state_sha256":"cc1ab495af5b5b8bc1e999549e477183a175be6fa1747e737de533beec37044f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nulyZjiAe1FDbRTZCSPWf4fho1xMYSax2DkyFgZ7qIqkBW1p1dfn+Zh8UAzgG5bAo6yhNYYvfxLrnc1MNvClAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T08:34:03.468566Z","bundle_sha256":"9b4d07758c44a736a2e497591a2c62e0a5cf57e93f8e487bf15800f053eaf8f4"}}