{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:GAPZRDJDPF6X5VWY67BAGAGLFL","short_pith_number":"pith:GAPZRDJD","canonical_record":{"source":{"id":"2605.05103","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-06T16:38:49Z","cross_cats_sorted":["cs.AI","cs.CY"],"title_canon_sha256":"4874ad54271ad6919cdd872c18fec828f92e02e123ba5af9d9bfc18341c73f0b","abstract_canon_sha256":"9de3296fb5908701f9e4331943c45674516bf4d682442327e2807beaeff9def5"},"schema_version":"1.0"},"canonical_sha256":"301f988d23797d7ed6d8f7c20300cb2ac7682c1080cc43d337399e510e83c1bb","source":{"kind":"arxiv","id":"2605.05103","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.05103","created_at":"2026-08-13T01:29:49Z"},{"alias_kind":"arxiv_version","alias_value":"2605.05103v3","created_at":"2026-08-13T01:29:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.05103","created_at":"2026-08-13T01:29:49Z"},{"alias_kind":"pith_short_12","alias_value":"GAPZRDJDPF6X","created_at":"2026-08-13T01:29:49Z"},{"alias_kind":"pith_short_16","alias_value":"GAPZRDJDPF6X5VWY","created_at":"2026-08-13T01:29:49Z"},{"alias_kind":"pith_short_8","alias_value":"GAPZRDJD","created_at":"2026-08-13T01:29:49Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:GAPZRDJDPF6X5VWY67BAGAGLFL","target":"record","payload":{"canonical_record":{"source":{"id":"2605.05103","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-06T16:38:49Z","cross_cats_sorted":["cs.AI","cs.CY"],"title_canon_sha256":"4874ad54271ad6919cdd872c18fec828f92e02e123ba5af9d9bfc18341c73f0b","abstract_canon_sha256":"9de3296fb5908701f9e4331943c45674516bf4d682442327e2807beaeff9def5"},"schema_version":"1.0"},"canonical_sha256":"301f988d23797d7ed6d8f7c20300cb2ac7682c1080cc43d337399e510e83c1bb","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-13T01:29:49.341937Z","signature_b64":"z4HyboVclT2iOiia1ZrOxs+f4uLl5+qBx+IcTtD7Ad9PUhTVnnXR4sRQE9Tsp9RsNHXohePRSmwlpuOc8oSVCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"301f988d23797d7ed6d8f7c20300cb2ac7682c1080cc43d337399e510e83c1bb","last_reissued_at":"2026-08-13T01:29:49.339728Z","signature_status":"signed_v1","first_computed_at":"2026-08-13T01:29:49.339728Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.05103","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":"2026-08-13T01:29:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"LYStCjcU4e/shTRob2YIbNfSKz96vLQvPfiIazE5rAcduBE40Ios6Bch09Coj0R/vVIe/yJmacGOlojM3WUKDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T19:11:45.970435Z"},"content_sha256":"d5ab71617610661cc55a1127ba35c42ca4324a60f31f3c8691ba00fd182b018f","schema_version":"1.0","event_id":"sha256:d5ab71617610661cc55a1127ba35c42ca4324a60f31f3c8691ba00fd182b018f"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:GAPZRDJDPF6X5VWY67BAGAGLFL","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Text Corpora as Concept Fields: Black-Box Hallucination and Novelty Measurement","license":"http://creativecommons.org/licenses/by/4.0/","headline":"A text corpus defines a concept field in sentence-embedding space whose local drift scores the groundedness of new sentences.","cross_cats":["cs.AI","cs.CY"],"primary_cat":"cs.CL","authors_text":"Chieh Ting Yeh, Khaoula Allak, Nicholas S. Kersting, Saad Taame, Vittorio Castelli, Xinzhu Wang","submitted_at":"2026-05-06T16:38:49Z","abstract_excerpt":"We introduce the \\textbf{Concept Field} of a text corpus: a local drift field with pointwise uncertainty, estimated in sentence-embedding space from the deltas between consecutive sentences. Given a candidate sentence transition, we score its agreement with the field by $\\zeta$, the mean absolute z-distance between the observed delta and the field's local Gaussian estimate. The score is black-box (no model internals), corpus-attributable (every score traces to nearby corpus sentences), and admits a probabilistically motivated interpretation under a local Gaussian approximation. We support the "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"On controlled LLM-generated rewrites, Concept Fields achieve strong selective classification performance under a grounded / ungrounded / unsure triage policy. Unlike retrieval-centric baselines, the resulting coverage-risk behavior is similar across both domains, supporting a degree of cross-domain stability for the standardized deviation score.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That deltas between consecutive sentences in sentence-embedding space admit a local Gaussian approximation whose z-distance yields a probabilistically meaningful groundedness score.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Concept Fields model text corpora as local Gaussian drift fields in embedding space to score sentence transitions for hallucination detection and novelty via standardized deviation.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A text corpus defines a concept field in sentence-embedding space whose local drift scores the groundedness of new sentences.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"4f1a4e58f5ba92c3b147fbeb9d91d2024a33c2687dbcd71287541afe7d7d22fe"},"source":{"id":"2605.05103","kind":"arxiv","version":3},"verdict":{"id":"d6448a33-4064-47ab-9e9a-0e3991d59ebe","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-12T03:39:17.349363Z","strongest_claim":"On controlled LLM-generated rewrites, Concept Fields achieve strong selective classification performance under a grounded / ungrounded / unsure triage policy. Unlike retrieval-centric baselines, the resulting coverage-risk behavior is similar across both domains, supporting a degree of cross-domain stability for the standardized deviation score.","one_line_summary":"Concept Fields model text corpora as local Gaussian drift fields in embedding space to score sentence transitions for hallucination detection and novelty via standardized deviation.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That deltas between consecutive sentences in sentence-embedding space admit a local Gaussian approximation whose z-distance yields a probabilistically meaningful groundedness score.","pith_extraction_headline":"A text corpus defines a concept field in sentence-embedding space whose local drift scores the groundedness of new sentences."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.05103/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-20T10:37:38.201620Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T21:31:19.590204Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T13:49:45.750063Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"e5610ba724e9b6fae90037edfc5a0a0d9956779e08e88c9f365fdb31ab380f0b"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"46056f568b0c6f84f79dad267b15bf05cf9c6a10d231532a80a82349a451108f"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"d6448a33-4064-47ab-9e9a-0e3991d59ebe"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-08-13T01:29:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"8a1esDVhNEOzNrobYxpRZiKVqXavotk/+5z8wRztx2qWRWI1P3GXF6JLXr9a723jmdmE1PfpdjumxB3ScdkcCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T19:11:45.971162Z"},"content_sha256":"479991927c477cc78dc13257132c7a7219c58ed73bfde44d61a80497b0f74976","schema_version":"1.0","event_id":"sha256:479991927c477cc78dc13257132c7a7219c58ed73bfde44d61a80497b0f74976"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/GAPZRDJDPF6X5VWY67BAGAGLFL/bundle.json","state_url":"https://pith.science/pith/GAPZRDJDPF6X5VWY67BAGAGLFL/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/GAPZRDJDPF6X5VWY67BAGAGLFL/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":"2026-08-14T19:11:45Z","links":{"resolver":"https://pith.science/pith/GAPZRDJDPF6X5VWY67BAGAGLFL","bundle":"https://pith.science/pith/GAPZRDJDPF6X5VWY67BAGAGLFL/bundle.json","state":"https://pith.science/pith/GAPZRDJDPF6X5VWY67BAGAGLFL/state.json","well_known_bundle":"https://pith.science/.well-known/pith/GAPZRDJDPF6X5VWY67BAGAGLFL/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:GAPZRDJDPF6X5VWY67BAGAGLFL","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":"9de3296fb5908701f9e4331943c45674516bf4d682442327e2807beaeff9def5","cross_cats_sorted":["cs.AI","cs.CY"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-06T16:38:49Z","title_canon_sha256":"4874ad54271ad6919cdd872c18fec828f92e02e123ba5af9d9bfc18341c73f0b"},"schema_version":"1.0","source":{"id":"2605.05103","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.05103","created_at":"2026-08-13T01:29:49Z"},{"alias_kind":"arxiv_version","alias_value":"2605.05103v3","created_at":"2026-08-13T01:29:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.05103","created_at":"2026-08-13T01:29:49Z"},{"alias_kind":"pith_short_12","alias_value":"GAPZRDJDPF6X","created_at":"2026-08-13T01:29:49Z"},{"alias_kind":"pith_short_16","alias_value":"GAPZRDJDPF6X5VWY","created_at":"2026-08-13T01:29:49Z"},{"alias_kind":"pith_short_8","alias_value":"GAPZRDJD","created_at":"2026-08-13T01:29:49Z"}],"graph_snapshots":[{"event_id":"sha256:479991927c477cc78dc13257132c7a7219c58ed73bfde44d61a80497b0f74976","target":"graph","created_at":"2026-08-13T01:29:49Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"On controlled LLM-generated rewrites, Concept Fields achieve strong selective classification performance under a grounded / ungrounded / unsure triage policy. Unlike retrieval-centric baselines, the resulting coverage-risk behavior is similar across both domains, supporting a degree of cross-domain stability for the standardized deviation score."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That deltas between consecutive sentences in sentence-embedding space admit a local Gaussian approximation whose z-distance yields a probabilistically meaningful groundedness score."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Concept Fields model text corpora as local Gaussian drift fields in embedding space to score sentence transitions for hallucination detection and novelty via standardized deviation."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"A text corpus defines a concept field in sentence-embedding space whose local drift scores the groundedness of new sentences."}],"snapshot_sha256":"4f1a4e58f5ba92c3b147fbeb9d91d2024a33c2687dbcd71287541afe7d7d22fe"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"46056f568b0c6f84f79dad267b15bf05cf9c6a10d231532a80a82349a451108f"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-20T10:37:38.201620Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-19T21:31:19.590204Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T13:49:45.750063Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2605.05103/integrity.json","findings":[],"snapshot_sha256":"e5610ba724e9b6fae90037edfc5a0a0d9956779e08e88c9f365fdb31ab380f0b","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We introduce the \\textbf{Concept Field} of a text corpus: a local drift field with pointwise uncertainty, estimated in sentence-embedding space from the deltas between consecutive sentences. Given a candidate sentence transition, we score its agreement with the field by $\\zeta$, the mean absolute z-distance between the observed delta and the field's local Gaussian estimate. The score is black-box (no model internals), corpus-attributable (every score traces to nearby corpus sentences), and admits a probabilistically motivated interpretation under a local Gaussian approximation. We support the ","authors_text":"Chieh Ting Yeh, Khaoula Allak, Nicholas S. Kersting, Saad Taame, Vittorio Castelli, Xinzhu Wang","cross_cats":["cs.AI","cs.CY"],"headline":"A text corpus defines a concept field in sentence-embedding space whose local drift scores the groundedness of new sentences.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-06T16:38:49Z","title":"Text Corpora as Concept Fields: Black-Box Hallucination and Novelty Measurement"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.05103","kind":"arxiv","version":3},"verdict":{"created_at":"2026-05-12T03:39:17.349363Z","id":"d6448a33-4064-47ab-9e9a-0e3991d59ebe","model_set":{"reader":"grok-4.3"},"one_line_summary":"Concept Fields model text corpora as local Gaussian drift fields in embedding space to score sentence transitions for hallucination detection and novelty via standardized deviation.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"A text corpus defines a concept field in sentence-embedding space whose local drift scores the groundedness of new sentences.","strongest_claim":"On controlled LLM-generated rewrites, Concept Fields achieve strong selective classification performance under a grounded / ungrounded / unsure triage policy. Unlike retrieval-centric baselines, the resulting coverage-risk behavior is similar across both domains, supporting a degree of cross-domain stability for the standardized deviation score.","weakest_assumption":"That deltas between consecutive sentences in sentence-embedding space admit a local Gaussian approximation whose z-distance yields a probabilistically meaningful groundedness score."}},"verdict_id":"d6448a33-4064-47ab-9e9a-0e3991d59ebe"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:d5ab71617610661cc55a1127ba35c42ca4324a60f31f3c8691ba00fd182b018f","target":"record","created_at":"2026-08-13T01:29:49Z","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":"9de3296fb5908701f9e4331943c45674516bf4d682442327e2807beaeff9def5","cross_cats_sorted":["cs.AI","cs.CY"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-06T16:38:49Z","title_canon_sha256":"4874ad54271ad6919cdd872c18fec828f92e02e123ba5af9d9bfc18341c73f0b"},"schema_version":"1.0","source":{"id":"2605.05103","kind":"arxiv","version":3}},"canonical_sha256":"301f988d23797d7ed6d8f7c20300cb2ac7682c1080cc43d337399e510e83c1bb","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"301f988d23797d7ed6d8f7c20300cb2ac7682c1080cc43d337399e510e83c1bb","first_computed_at":"2026-08-13T01:29:49.339728Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-08-13T01:29:49.339728Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"z4HyboVclT2iOiia1ZrOxs+f4uLl5+qBx+IcTtD7Ad9PUhTVnnXR4sRQE9Tsp9RsNHXohePRSmwlpuOc8oSVCQ==","signature_status":"signed_v1","signed_at":"2026-08-13T01:29:49.341937Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.05103","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d5ab71617610661cc55a1127ba35c42ca4324a60f31f3c8691ba00fd182b018f","sha256:479991927c477cc78dc13257132c7a7219c58ed73bfde44d61a80497b0f74976"],"state_sha256":"d464c6e90cfb0c0d620890eef4d9258f50f5b1865c403e4cf5ef5dec41e48f17"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"b3l/LVwzlhOVChN5kfy/Yl1hPfob4UwJVNqUguW3QaFxEXdOph8acQ9tO81Q5tSSQkaZLQWLwOvfR0RuO1/yDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-14T19:11:45.975753Z","bundle_sha256":"b87035c9b2865b37d4105ba79325c19c96a8708f800c58300383ad32d48e6a11"}}