{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:343Y7JVWDOITES3AUVDVOK4PCB","short_pith_number":"pith:343Y7JVW","canonical_record":{"source":{"id":"2505.00506","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-01T13:22:45Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"7d99be9d9dcfc6bdf3e69a53cf7bd3ebcac0d1b3f2805c502b03848979704e1f","abstract_canon_sha256":"31cf8ac75c4a14ab4157158ed058a7a49c18fe4bd46edc82455ea1309c5503fb"},"schema_version":"1.0"},"canonical_sha256":"df378fa6b61b91324b60a547572b8f106c8e583578e2f7410834a4b4a6216ec0","source":{"kind":"arxiv","id":"2505.00506","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.00506","created_at":"2026-07-05T10:57:14Z"},{"alias_kind":"arxiv_version","alias_value":"2505.00506v1","created_at":"2026-07-05T10:57:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.00506","created_at":"2026-07-05T10:57:14Z"},{"alias_kind":"pith_short_12","alias_value":"343Y7JVWDOIT","created_at":"2026-07-05T10:57:14Z"},{"alias_kind":"pith_short_16","alias_value":"343Y7JVWDOITES3A","created_at":"2026-07-05T10:57:14Z"},{"alias_kind":"pith_short_8","alias_value":"343Y7JVW","created_at":"2026-07-05T10:57:14Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:343Y7JVWDOITES3AUVDVOK4PCB","target":"record","payload":{"canonical_record":{"source":{"id":"2505.00506","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-01T13:22:45Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"7d99be9d9dcfc6bdf3e69a53cf7bd3ebcac0d1b3f2805c502b03848979704e1f","abstract_canon_sha256":"31cf8ac75c4a14ab4157158ed058a7a49c18fe4bd46edc82455ea1309c5503fb"},"schema_version":"1.0"},"canonical_sha256":"df378fa6b61b91324b60a547572b8f106c8e583578e2f7410834a4b4a6216ec0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:57:14.666013Z","signature_b64":"7TQ9tD4pieuVk3plibvEsBu7R+TP1A3jy5p0BYLFLD+72ZcIvG1KxAyu2+K+nBWadtDXs/3M1guxcTCxZxoEAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"df378fa6b61b91324b60a547572b8f106c8e583578e2f7410834a4b4a6216ec0","last_reissued_at":"2026-07-05T10:57:14.665586Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:57:14.665586Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.00506","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":"2026-07-05T10:57:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JMWX4LWyIY3doAGx4bJDZymh6jHdPxntClbspXc7roPFh2POqNjjvXateGssJKgEhL8kmNjfNxHW5JTSECGNCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T06:00:28.837125Z"},"content_sha256":"2b7c1616f7411b7c86dfb3c5b2cbd1543e4fe6a2d3f88f9c876579de030593c0","schema_version":"1.0","event_id":"sha256:2b7c1616f7411b7c86dfb3c5b2cbd1543e4fe6a2d3f88f9c876579de030593c0"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:343Y7JVWDOITES3AUVDVOK4PCB","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"HalluMix: A Task-Agnostic, Multi-Domain Benchmark for Real-World Hallucination Detection","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Deanna Emery, Freddie Vargus, Iulia Neagu, Michael Goitia","submitted_at":"2025-05-01T13:22:45Z","abstract_excerpt":"As large language models (LLMs) are increasingly deployed in high-stakes domains, detecting hallucinated content$\\unicode{x2013}$text that is not grounded in supporting evidence$\\unicode{x2013}$has become a critical challenge. Existing benchmarks for hallucination detection are often synthetically generated, narrowly focused on extractive question answering, and fail to capture the complexity of real-world scenarios involving multi-document contexts and full-sentence outputs. We introduce the HalluMix Benchmark, a diverse, task-agnostic dataset that includes examples from a range of domains an"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.00506","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/2505.00506/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":"2026-07-05T10:57:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"jzR8AfebfxX9NrGs9i0OF3DsBFD3zWMP1mQ73wtM65csd0/J6LKSo3OZjO0NDRVkoTmde7KyWSLBEtm6ujgyDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-18T06:00:28.837622Z"},"content_sha256":"85e76582d99b5b99232623ebe8ffe880ed515586915346a1141e3a1e89d95020","schema_version":"1.0","event_id":"sha256:85e76582d99b5b99232623ebe8ffe880ed515586915346a1141e3a1e89d95020"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/343Y7JVWDOITES3AUVDVOK4PCB/bundle.json","state_url":"https://pith.science/pith/343Y7JVWDOITES3AUVDVOK4PCB/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/343Y7JVWDOITES3AUVDVOK4PCB/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":"2026-08-18T06:00:28Z","links":{"resolver":"https://pith.science/pith/343Y7JVWDOITES3AUVDVOK4PCB","bundle":"https://pith.science/pith/343Y7JVWDOITES3AUVDVOK4PCB/bundle.json","state":"https://pith.science/pith/343Y7JVWDOITES3AUVDVOK4PCB/state.json","well_known_bundle":"https://pith.science/.well-known/pith/343Y7JVWDOITES3AUVDVOK4PCB/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:343Y7JVWDOITES3AUVDVOK4PCB","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":"31cf8ac75c4a14ab4157158ed058a7a49c18fe4bd46edc82455ea1309c5503fb","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-01T13:22:45Z","title_canon_sha256":"7d99be9d9dcfc6bdf3e69a53cf7bd3ebcac0d1b3f2805c502b03848979704e1f"},"schema_version":"1.0","source":{"id":"2505.00506","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.00506","created_at":"2026-07-05T10:57:14Z"},{"alias_kind":"arxiv_version","alias_value":"2505.00506v1","created_at":"2026-07-05T10:57:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.00506","created_at":"2026-07-05T10:57:14Z"},{"alias_kind":"pith_short_12","alias_value":"343Y7JVWDOIT","created_at":"2026-07-05T10:57:14Z"},{"alias_kind":"pith_short_16","alias_value":"343Y7JVWDOITES3A","created_at":"2026-07-05T10:57:14Z"},{"alias_kind":"pith_short_8","alias_value":"343Y7JVW","created_at":"2026-07-05T10:57:14Z"}],"graph_snapshots":[{"event_id":"sha256:85e76582d99b5b99232623ebe8ffe880ed515586915346a1141e3a1e89d95020","target":"graph","created_at":"2026-07-05T10:57:14Z","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/2505.00506/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As large language models (LLMs) are increasingly deployed in high-stakes domains, detecting hallucinated content$\\unicode{x2013}$text that is not grounded in supporting evidence$\\unicode{x2013}$has become a critical challenge. Existing benchmarks for hallucination detection are often synthetically generated, narrowly focused on extractive question answering, and fail to capture the complexity of real-world scenarios involving multi-document contexts and full-sentence outputs. We introduce the HalluMix Benchmark, a diverse, task-agnostic dataset that includes examples from a range of domains an","authors_text":"Deanna Emery, Freddie Vargus, Iulia Neagu, Michael Goitia","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-01T13:22:45Z","title":"HalluMix: A Task-Agnostic, Multi-Domain Benchmark for Real-World Hallucination Detection"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.00506","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":"sha256:2b7c1616f7411b7c86dfb3c5b2cbd1543e4fe6a2d3f88f9c876579de030593c0","target":"record","created_at":"2026-07-05T10:57:14Z","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":"31cf8ac75c4a14ab4157158ed058a7a49c18fe4bd46edc82455ea1309c5503fb","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-01T13:22:45Z","title_canon_sha256":"7d99be9d9dcfc6bdf3e69a53cf7bd3ebcac0d1b3f2805c502b03848979704e1f"},"schema_version":"1.0","source":{"id":"2505.00506","kind":"arxiv","version":1}},"canonical_sha256":"df378fa6b61b91324b60a547572b8f106c8e583578e2f7410834a4b4a6216ec0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"df378fa6b61b91324b60a547572b8f106c8e583578e2f7410834a4b4a6216ec0","first_computed_at":"2026-07-05T10:57:14.665586Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:57:14.665586Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"7TQ9tD4pieuVk3plibvEsBu7R+TP1A3jy5p0BYLFLD+72ZcIvG1KxAyu2+K+nBWadtDXs/3M1guxcTCxZxoEAg==","signature_status":"signed_v1","signed_at":"2026-07-05T10:57:14.666013Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.00506","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2b7c1616f7411b7c86dfb3c5b2cbd1543e4fe6a2d3f88f9c876579de030593c0","sha256:85e76582d99b5b99232623ebe8ffe880ed515586915346a1141e3a1e89d95020"],"state_sha256":"a9d538f50e2d25403c5d122edd8b9fac13df259e2eca5a55715bcfefe541ee52"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ARRWKV4YcKQGw2VvhS6ANbdZkU6Yn5Cbvjkjd1hb0iE5NQsldsjJSz3PjxGWhRCz9SGJrVnS9NK7Kj5NZwmSAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-18T06:00:28.841388Z","bundle_sha256":"f67267991f47f295bb618a35a947af97d2011766942325ec81a4875e7d40ae80"}}