{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:KLMBXVYKV4QV6H5KV5NX5PTZAC","short_pith_number":"pith:KLMBXVYK","canonical_record":{"source":{"id":"2305.14552","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-05-23T22:24:44Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2a21c4d5476faf9310aa847660c9e4b383599ea205be040e06bfd5133b0d4977","abstract_canon_sha256":"eb9e0d4494b3c3a326ab3b336e424b225789421eb9e8dbbc109c75f5d9288d9b"},"schema_version":"1.0"},"canonical_sha256":"52d81bd70aaf215f1faaaf5b7ebe7900b66d900164bb785d79db7be8bdfc9414","source":{"kind":"arxiv","id":"2305.14552","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2305.14552","created_at":"2026-07-05T07:03:25Z"},{"alias_kind":"arxiv_version","alias_value":"2305.14552v2","created_at":"2026-07-05T07:03:25Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.14552","created_at":"2026-07-05T07:03:25Z"},{"alias_kind":"pith_short_12","alias_value":"KLMBXVYKV4QV","created_at":"2026-07-05T07:03:25Z"},{"alias_kind":"pith_short_16","alias_value":"KLMBXVYKV4QV6H5K","created_at":"2026-07-05T07:03:25Z"},{"alias_kind":"pith_short_8","alias_value":"KLMBXVYK","created_at":"2026-07-05T07:03:25Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:KLMBXVYKV4QV6H5KV5NX5PTZAC","target":"record","payload":{"canonical_record":{"source":{"id":"2305.14552","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-05-23T22:24:44Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2a21c4d5476faf9310aa847660c9e4b383599ea205be040e06bfd5133b0d4977","abstract_canon_sha256":"eb9e0d4494b3c3a326ab3b336e424b225789421eb9e8dbbc109c75f5d9288d9b"},"schema_version":"1.0"},"canonical_sha256":"52d81bd70aaf215f1faaaf5b7ebe7900b66d900164bb785d79db7be8bdfc9414","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:03:25.524182Z","signature_b64":"RNU7Rny7wrHyOFZR3n/Ca4mt//I4TILzE8zwbSnxg9VxI4BrNZsm4VgHwdjKoiYEyVhrOvp87FcOGWSMKzV2AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"52d81bd70aaf215f1faaaf5b7ebe7900b66d900164bb785d79db7be8bdfc9414","last_reissued_at":"2026-07-05T07:03:25.523698Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:03:25.523698Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2305.14552","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:03:25Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"c8Af2PR8NrYiu1LylZpYZd4YA07g+CbjpSD4D86ADQQqESkKfnN/ndlsIP0aLpp+Z555+I3NNa0miO85rOZyBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-12T13:07:18.477688Z"},"content_sha256":"b82770a2fc9a798dc52268417c04ee70b0f9074480edc5d709a8e43995954710","schema_version":"1.0","event_id":"sha256:b82770a2fc9a798dc52268417c04ee70b0f9074480edc5d709a8e43995954710"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:KLMBXVYKV4QV6H5KV5NX5PTZAC","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Sources of Hallucination by Large Language Models on Inference Tasks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Liang Cheng, Mark Johnson, Mark Steedman, Mohammad Javad Hosseini, Nick McKenna, Tianyi Li","submitted_at":"2023-05-23T22:24:44Z","abstract_excerpt":"Large Language Models (LLMs) are claimed to be capable of Natural Language Inference (NLI), necessary for applied tasks like question answering and summarization. We present a series of behavioral studies on several LLM families (LLaMA, GPT-3.5, and PaLM) which probe their behavior using controlled experiments. We establish two biases originating from pretraining which predict much of their behavior, and show that these are major sources of hallucination in generative LLMs. First, memorization at the level of sentences: we show that, regardless of the premise, models falsely label NLI test sam"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.14552","kind":"arxiv","version":2},"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/2305.14552/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":"2026-07-05T07:03:25Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Xqlq8LtszBo2KNItLXpm9vJHZPaXoTsqycRAErKnzaggTxD8YPmv81WnG+wmGaC/pcFHkhVLSo38E8VmRtluBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-12T13:07:18.478202Z"},"content_sha256":"8cfd5d1c5d15a1a915ab64655fc8a431744e51bdd1bfebe937d28af4809a5dcb","schema_version":"1.0","event_id":"sha256:8cfd5d1c5d15a1a915ab64655fc8a431744e51bdd1bfebe937d28af4809a5dcb"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KLMBXVYKV4QV6H5KV5NX5PTZAC/bundle.json","state_url":"https://pith.science/pith/KLMBXVYKV4QV6H5KV5NX5PTZAC/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KLMBXVYKV4QV6H5KV5NX5PTZAC/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":"2026-08-12T13:07:18Z","links":{"resolver":"https://pith.science/pith/KLMBXVYKV4QV6H5KV5NX5PTZAC","bundle":"https://pith.science/pith/KLMBXVYKV4QV6H5KV5NX5PTZAC/bundle.json","state":"https://pith.science/pith/KLMBXVYKV4QV6H5KV5NX5PTZAC/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KLMBXVYKV4QV6H5KV5NX5PTZAC/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:KLMBXVYKV4QV6H5KV5NX5PTZAC","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":"eb9e0d4494b3c3a326ab3b336e424b225789421eb9e8dbbc109c75f5d9288d9b","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-05-23T22:24:44Z","title_canon_sha256":"2a21c4d5476faf9310aa847660c9e4b383599ea205be040e06bfd5133b0d4977"},"schema_version":"1.0","source":{"id":"2305.14552","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2305.14552","created_at":"2026-07-05T07:03:25Z"},{"alias_kind":"arxiv_version","alias_value":"2305.14552v2","created_at":"2026-07-05T07:03:25Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.14552","created_at":"2026-07-05T07:03:25Z"},{"alias_kind":"pith_short_12","alias_value":"KLMBXVYKV4QV","created_at":"2026-07-05T07:03:25Z"},{"alias_kind":"pith_short_16","alias_value":"KLMBXVYKV4QV6H5K","created_at":"2026-07-05T07:03:25Z"},{"alias_kind":"pith_short_8","alias_value":"KLMBXVYK","created_at":"2026-07-05T07:03:25Z"}],"graph_snapshots":[{"event_id":"sha256:8cfd5d1c5d15a1a915ab64655fc8a431744e51bdd1bfebe937d28af4809a5dcb","target":"graph","created_at":"2026-07-05T07:03:25Z","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/2305.14552/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Models (LLMs) are claimed to be capable of Natural Language Inference (NLI), necessary for applied tasks like question answering and summarization. We present a series of behavioral studies on several LLM families (LLaMA, GPT-3.5, and PaLM) which probe their behavior using controlled experiments. We establish two biases originating from pretraining which predict much of their behavior, and show that these are major sources of hallucination in generative LLMs. First, memorization at the level of sentences: we show that, regardless of the premise, models falsely label NLI test sam","authors_text":"Liang Cheng, Mark Johnson, Mark Steedman, Mohammad Javad Hosseini, Nick McKenna, Tianyi Li","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-05-23T22:24:44Z","title":"Sources of Hallucination by Large Language Models on Inference Tasks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.14552","kind":"arxiv","version":2},"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:b82770a2fc9a798dc52268417c04ee70b0f9074480edc5d709a8e43995954710","target":"record","created_at":"2026-07-05T07:03:25Z","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":"eb9e0d4494b3c3a326ab3b336e424b225789421eb9e8dbbc109c75f5d9288d9b","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-05-23T22:24:44Z","title_canon_sha256":"2a21c4d5476faf9310aa847660c9e4b383599ea205be040e06bfd5133b0d4977"},"schema_version":"1.0","source":{"id":"2305.14552","kind":"arxiv","version":2}},"canonical_sha256":"52d81bd70aaf215f1faaaf5b7ebe7900b66d900164bb785d79db7be8bdfc9414","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"52d81bd70aaf215f1faaaf5b7ebe7900b66d900164bb785d79db7be8bdfc9414","first_computed_at":"2026-07-05T07:03:25.523698Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:03:25.523698Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"RNU7Rny7wrHyOFZR3n/Ca4mt//I4TILzE8zwbSnxg9VxI4BrNZsm4VgHwdjKoiYEyVhrOvp87FcOGWSMKzV2AA==","signature_status":"signed_v1","signed_at":"2026-07-05T07:03:25.524182Z","signed_message":"canonical_sha256_bytes"},"source_id":"2305.14552","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b82770a2fc9a798dc52268417c04ee70b0f9074480edc5d709a8e43995954710","sha256:8cfd5d1c5d15a1a915ab64655fc8a431744e51bdd1bfebe937d28af4809a5dcb"],"state_sha256":"85c62df3483421b922ac712984c5f6d248bce2c56f7baed4f7aff24720fcedd2"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"vZZfcxazRBvv2n3bghhIWnqUDe7909aFVtiPlu3jlT526nO4G3PfNAv59cHBDbXmF/93+OxJ6bAMTKE1prLzAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-12T13:07:18.482220Z","bundle_sha256":"df96dd7c05cfc860c9802b1db7880fc70642979f32a106d0e3e59c62900efa84"}}