{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:EGTA4JNRACN2GOKVSE3YTGXUXB","short_pith_number":"pith:EGTA4JNR","canonical_record":{"source":{"id":"2509.17932","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-22T15:54:29Z","cross_cats_sorted":[],"title_canon_sha256":"652fe37515e14f8038b4b99a5c0ead76e42bcacb8ba061d6759b3aa364baffca","abstract_canon_sha256":"cf9ab3e43e83750faa59774965514324a3dafadb840b317cb81182575b50eb86"},"schema_version":"1.0"},"canonical_sha256":"21a60e25b1009ba339559137899af4b84cd66db0bd179e0466ecebc75df001e3","source":{"kind":"arxiv","id":"2509.17932","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.17932","created_at":"2026-06-29T01:14:24Z"},{"alias_kind":"arxiv_version","alias_value":"2509.17932v2","created_at":"2026-06-29T01:14:24Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.17932","created_at":"2026-06-29T01:14:24Z"},{"alias_kind":"pith_short_12","alias_value":"EGTA4JNRACN2","created_at":"2026-06-29T01:14:24Z"},{"alias_kind":"pith_short_16","alias_value":"EGTA4JNRACN2GOKV","created_at":"2026-06-29T01:14:24Z"},{"alias_kind":"pith_short_8","alias_value":"EGTA4JNR","created_at":"2026-06-29T01:14:24Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:EGTA4JNRACN2GOKVSE3YTGXUXB","target":"record","payload":{"canonical_record":{"source":{"id":"2509.17932","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-22T15:54:29Z","cross_cats_sorted":[],"title_canon_sha256":"652fe37515e14f8038b4b99a5c0ead76e42bcacb8ba061d6759b3aa364baffca","abstract_canon_sha256":"cf9ab3e43e83750faa59774965514324a3dafadb840b317cb81182575b50eb86"},"schema_version":"1.0"},"canonical_sha256":"21a60e25b1009ba339559137899af4b84cd66db0bd179e0466ecebc75df001e3","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-29T01:14:24.204600Z","signature_b64":"eN1pzjytYDgTLGyb1F3lZx4/pvITBqpNsbpbYa4zU1mqbCFMN34yL61D5VSk/QT+5LNFT8y4KGd2RK5J01qACQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"21a60e25b1009ba339559137899af4b84cd66db0bd179e0466ecebc75df001e3","last_reissued_at":"2026-06-29T01:14:24.204123Z","signature_status":"signed_v1","first_computed_at":"2026-06-29T01:14:24.204123Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2509.17932","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":"2026-06-29T01:14:24Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"R24rCjF2ZWiWG5k+S2sgAzZUIiSF2hZtxfRwgAToaTRKzGiEQ0bZfVbLAjf0/LL6/veIDDG2AoaeM6xuKNI9Ag==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T21:22:59.390731Z"},"content_sha256":"887f293e6bae91388c22343c6c48982a92d894989701a710ea8c909810cb3d26","schema_version":"1.0","event_id":"sha256:887f293e6bae91388c22343c6c48982a92d894989701a710ea8c909810cb3d26"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:EGTA4JNRACN2GOKVSE3YTGXUXB","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Training-free Truthfulness Detection via Sparse MLP Value Vectors","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Heyan Huang, Runheng Liu, Xingchen Xiao, Yanghao Zhou, Zhijing Wu","submitted_at":"2025-09-22T15:54:29Z","abstract_excerpt":"Large language models (LLMs) are prone to generating factually incorrect content, motivating methods for assessing truthfulness from internal model signals. While supervised probing approaches can be effective, they require labeled data and classifier training. Recent training-free methods avoid parameter optimization but rely on coarse activation statistics that provide limited insight into how truthfulness-related signals arise within the model. We present a training-free approach that operates at the level of individual multi-layer perceptron (MLP) value vectors. Through a systematic analys"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.17932","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/2509.17932/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":"2026-06-29T01:14:24Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"d3y7hQBaPsdOQKwmtjqIUXWb0hiIBlU2tQ87JObxGotpjUjuIrD2i6kGf900vc5UmXPfKoCDqycMAsSLFqvIDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T21:22:59.391643Z"},"content_sha256":"c765fae76eca4684e5bbcec32575d1376ca1692fa35070113addcd33ccf44175","schema_version":"1.0","event_id":"sha256:c765fae76eca4684e5bbcec32575d1376ca1692fa35070113addcd33ccf44175"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/EGTA4JNRACN2GOKVSE3YTGXUXB/bundle.json","state_url":"https://pith.science/pith/EGTA4JNRACN2GOKVSE3YTGXUXB/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/EGTA4JNRACN2GOKVSE3YTGXUXB/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":"2026-08-09T21:22:59Z","links":{"resolver":"https://pith.science/pith/EGTA4JNRACN2GOKVSE3YTGXUXB","bundle":"https://pith.science/pith/EGTA4JNRACN2GOKVSE3YTGXUXB/bundle.json","state":"https://pith.science/pith/EGTA4JNRACN2GOKVSE3YTGXUXB/state.json","well_known_bundle":"https://pith.science/.well-known/pith/EGTA4JNRACN2GOKVSE3YTGXUXB/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:EGTA4JNRACN2GOKVSE3YTGXUXB","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":"cf9ab3e43e83750faa59774965514324a3dafadb840b317cb81182575b50eb86","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-22T15:54:29Z","title_canon_sha256":"652fe37515e14f8038b4b99a5c0ead76e42bcacb8ba061d6759b3aa364baffca"},"schema_version":"1.0","source":{"id":"2509.17932","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2509.17932","created_at":"2026-06-29T01:14:24Z"},{"alias_kind":"arxiv_version","alias_value":"2509.17932v2","created_at":"2026-06-29T01:14:24Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.17932","created_at":"2026-06-29T01:14:24Z"},{"alias_kind":"pith_short_12","alias_value":"EGTA4JNRACN2","created_at":"2026-06-29T01:14:24Z"},{"alias_kind":"pith_short_16","alias_value":"EGTA4JNRACN2GOKV","created_at":"2026-06-29T01:14:24Z"},{"alias_kind":"pith_short_8","alias_value":"EGTA4JNR","created_at":"2026-06-29T01:14:24Z"}],"graph_snapshots":[{"event_id":"sha256:c765fae76eca4684e5bbcec32575d1376ca1692fa35070113addcd33ccf44175","target":"graph","created_at":"2026-06-29T01:14:24Z","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/2509.17932/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) are prone to generating factually incorrect content, motivating methods for assessing truthfulness from internal model signals. While supervised probing approaches can be effective, they require labeled data and classifier training. Recent training-free methods avoid parameter optimization but rely on coarse activation statistics that provide limited insight into how truthfulness-related signals arise within the model. We present a training-free approach that operates at the level of individual multi-layer perceptron (MLP) value vectors. Through a systematic analys","authors_text":"Heyan Huang, Runheng Liu, Xingchen Xiao, Yanghao Zhou, Zhijing Wu","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-22T15:54:29Z","title":"Training-free Truthfulness Detection via Sparse MLP Value Vectors"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.17932","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":"sha256:887f293e6bae91388c22343c6c48982a92d894989701a710ea8c909810cb3d26","target":"record","created_at":"2026-06-29T01:14:24Z","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":"cf9ab3e43e83750faa59774965514324a3dafadb840b317cb81182575b50eb86","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-22T15:54:29Z","title_canon_sha256":"652fe37515e14f8038b4b99a5c0ead76e42bcacb8ba061d6759b3aa364baffca"},"schema_version":"1.0","source":{"id":"2509.17932","kind":"arxiv","version":2}},"canonical_sha256":"21a60e25b1009ba339559137899af4b84cd66db0bd179e0466ecebc75df001e3","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"21a60e25b1009ba339559137899af4b84cd66db0bd179e0466ecebc75df001e3","first_computed_at":"2026-06-29T01:14:24.204123Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-29T01:14:24.204123Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"eN1pzjytYDgTLGyb1F3lZx4/pvITBqpNsbpbYa4zU1mqbCFMN34yL61D5VSk/QT+5LNFT8y4KGd2RK5J01qACQ==","signature_status":"signed_v1","signed_at":"2026-06-29T01:14:24.204600Z","signed_message":"canonical_sha256_bytes"},"source_id":"2509.17932","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:887f293e6bae91388c22343c6c48982a92d894989701a710ea8c909810cb3d26","sha256:c765fae76eca4684e5bbcec32575d1376ca1692fa35070113addcd33ccf44175"],"state_sha256":"88f1888143fa206e5aa18d4828ddb5d27dde431c835e3c7822eb98bb9a87d604"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6B6+eh4KG4DLfo4G6ZIq2U0xmA2UtUZSgTcIx4V8QwAn6XJEoKqxGMT3LhkasC6t6YLNmFpmU0q4oIlFG6DiAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T21:22:59.412631Z","bundle_sha256":"1cec21f7420ac8f4a5f07657b76667fdf7ea771b9b054d961ed15267f10c561e"}}