{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:FWJRLGLGVKT52HJRBIHANRCIA6","short_pith_number":"pith:FWJRLGLG","canonical_record":{"source":{"id":"2508.00901","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-07-28T17:24:57Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"699e697e29a21dd2dc72d8dd8abc551b2ca412654d866dbe1768e571001a82ca","abstract_canon_sha256":"5bb133b7956eb57f5e3faa08133607546118df9835cca9b4365850d10d03bf6d"},"schema_version":"1.0"},"canonical_sha256":"2d93159966aaa7dd1d310a0e06c4480792ea1d30e9dbd6b97f11868805b6de14","source":{"kind":"arxiv","id":"2508.00901","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.00901","created_at":"2026-05-20T00:05:31Z"},{"alias_kind":"arxiv_version","alias_value":"2508.00901v4","created_at":"2026-05-20T00:05:31Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.00901","created_at":"2026-05-20T00:05:31Z"},{"alias_kind":"pith_short_12","alias_value":"FWJRLGLGVKT5","created_at":"2026-05-20T00:05:31Z"},{"alias_kind":"pith_short_16","alias_value":"FWJRLGLGVKT52HJR","created_at":"2026-05-20T00:05:31Z"},{"alias_kind":"pith_short_8","alias_value":"FWJRLGLG","created_at":"2026-05-20T00:05:31Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:FWJRLGLGVKT52HJRBIHANRCIA6","target":"record","payload":{"canonical_record":{"source":{"id":"2508.00901","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-07-28T17:24:57Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"699e697e29a21dd2dc72d8dd8abc551b2ca412654d866dbe1768e571001a82ca","abstract_canon_sha256":"5bb133b7956eb57f5e3faa08133607546118df9835cca9b4365850d10d03bf6d"},"schema_version":"1.0"},"canonical_sha256":"2d93159966aaa7dd1d310a0e06c4480792ea1d30e9dbd6b97f11868805b6de14","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:05:31.994752Z","signature_b64":"3pLdy2lca9Z1CsVpkWllx4Ci1rGd/8GKeje41gELTa2MwVMZPJHXCbz0bDdrnUUi3vHXoAyw48a9CGJ3NKlwAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2d93159966aaa7dd1d310a0e06c4480792ea1d30e9dbd6b97f11868805b6de14","last_reissued_at":"2026-05-20T00:05:31.993886Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:05:31.993886Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2508.00901","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T00:05:31Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"i8691yGDRxAMdmkPIiTSga6VeJbmDXg2ui5haB+SuZzYpTEllPqfr/FLDFKsZb6DsnIHnNBrXwM4qRjmYas1DQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T09:42:48.881548Z"},"content_sha256":"568060a22a9fed14c60f3b5d9832f16e3a03a33627f46825b44f2052a779fa55","schema_version":"1.0","event_id":"sha256:568060a22a9fed14c60f3b5d9832f16e3a03a33627f46825b44f2052a779fa55"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:FWJRLGLGVKT52HJRBIHANRCIA6","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Provable Knowledge Acquisition and Extraction in One-Layer Transformers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Kexin Chen, Ruichen Xu","submitted_at":"2025-07-28T17:24:57Z","abstract_excerpt":"Large language models may encounter factual knowledge during pre-training yet fail to reliably use that knowledge after fine-tuning. Despite growing empirical evidence that MLP layers store factual associations and fine-tuning affects factual recall, the training-dynamics mechanisms linking next-token pre-training, knowledge storage, and post-fine-tuning extraction remain poorly understood. We study this problem in a stylized one-layer transformer with self-attention and MLP modules, trained by next-token prediction and subsequently fine-tuned on question-answering data. Under suitable regular"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.00901","kind":"arxiv","version":4},"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/2508.00901/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":"2026-05-20T00:05:31Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"w6PF+rkTxIgyhR401sGEdagjYhlUthaV7Yv0KA0mYPhHm8jePdLv4GcNV8yhtjm84SsYPQWG3Qdc5LHM+HhABA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T09:42:48.882272Z"},"content_sha256":"f87533932460508bc1eec3e3d13b8214fcfdac0ae59d0941876f1954928add63","schema_version":"1.0","event_id":"sha256:f87533932460508bc1eec3e3d13b8214fcfdac0ae59d0941876f1954928add63"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/FWJRLGLGVKT52HJRBIHANRCIA6/bundle.json","state_url":"https://pith.science/pith/FWJRLGLGVKT52HJRBIHANRCIA6/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/FWJRLGLGVKT52HJRBIHANRCIA6/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":"2026-05-26T09:42:48Z","links":{"resolver":"https://pith.science/pith/FWJRLGLGVKT52HJRBIHANRCIA6","bundle":"https://pith.science/pith/FWJRLGLGVKT52HJRBIHANRCIA6/bundle.json","state":"https://pith.science/pith/FWJRLGLGVKT52HJRBIHANRCIA6/state.json","well_known_bundle":"https://pith.science/.well-known/pith/FWJRLGLGVKT52HJRBIHANRCIA6/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:FWJRLGLGVKT52HJRBIHANRCIA6","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":"5bb133b7956eb57f5e3faa08133607546118df9835cca9b4365850d10d03bf6d","cross_cats_sorted":["cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-07-28T17:24:57Z","title_canon_sha256":"699e697e29a21dd2dc72d8dd8abc551b2ca412654d866dbe1768e571001a82ca"},"schema_version":"1.0","source":{"id":"2508.00901","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.00901","created_at":"2026-05-20T00:05:31Z"},{"alias_kind":"arxiv_version","alias_value":"2508.00901v4","created_at":"2026-05-20T00:05:31Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.00901","created_at":"2026-05-20T00:05:31Z"},{"alias_kind":"pith_short_12","alias_value":"FWJRLGLGVKT5","created_at":"2026-05-20T00:05:31Z"},{"alias_kind":"pith_short_16","alias_value":"FWJRLGLGVKT52HJR","created_at":"2026-05-20T00:05:31Z"},{"alias_kind":"pith_short_8","alias_value":"FWJRLGLG","created_at":"2026-05-20T00:05:31Z"}],"graph_snapshots":[{"event_id":"sha256:f87533932460508bc1eec3e3d13b8214fcfdac0ae59d0941876f1954928add63","target":"graph","created_at":"2026-05-20T00:05:31Z","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/2508.00901/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models may encounter factual knowledge during pre-training yet fail to reliably use that knowledge after fine-tuning. Despite growing empirical evidence that MLP layers store factual associations and fine-tuning affects factual recall, the training-dynamics mechanisms linking next-token pre-training, knowledge storage, and post-fine-tuning extraction remain poorly understood. We study this problem in a stylized one-layer transformer with self-attention and MLP modules, trained by next-token prediction and subsequently fine-tuned on question-answering data. Under suitable regular","authors_text":"Kexin Chen, Ruichen Xu","cross_cats":["cs.CL"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-07-28T17:24:57Z","title":"Provable Knowledge Acquisition and Extraction in One-Layer Transformers"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.00901","kind":"arxiv","version":4},"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":"sha256:568060a22a9fed14c60f3b5d9832f16e3a03a33627f46825b44f2052a779fa55","target":"record","created_at":"2026-05-20T00:05:31Z","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":"5bb133b7956eb57f5e3faa08133607546118df9835cca9b4365850d10d03bf6d","cross_cats_sorted":["cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-07-28T17:24:57Z","title_canon_sha256":"699e697e29a21dd2dc72d8dd8abc551b2ca412654d866dbe1768e571001a82ca"},"schema_version":"1.0","source":{"id":"2508.00901","kind":"arxiv","version":4}},"canonical_sha256":"2d93159966aaa7dd1d310a0e06c4480792ea1d30e9dbd6b97f11868805b6de14","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2d93159966aaa7dd1d310a0e06c4480792ea1d30e9dbd6b97f11868805b6de14","first_computed_at":"2026-05-20T00:05:31.993886Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:05:31.993886Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"3pLdy2lca9Z1CsVpkWllx4Ci1rGd/8GKeje41gELTa2MwVMZPJHXCbz0bDdrnUUi3vHXoAyw48a9CGJ3NKlwAg==","signature_status":"signed_v1","signed_at":"2026-05-20T00:05:31.994752Z","signed_message":"canonical_sha256_bytes"},"source_id":"2508.00901","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:568060a22a9fed14c60f3b5d9832f16e3a03a33627f46825b44f2052a779fa55","sha256:f87533932460508bc1eec3e3d13b8214fcfdac0ae59d0941876f1954928add63"],"state_sha256":"531e2a241ef8097bad31434eb1886819868ce229a98f095252a9f12c1db38cad"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qeFb/1FcqY/p4lH9VNFCWSySKljx/bMyKxcG3u6thEQN6ykWYZ+8mMJh/yfZU8kCzKSm5RVgdZ0UvkhLs6A3Dw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-26T09:42:48.886292Z","bundle_sha256":"a9cbc45e10e2fa944b5081fc698cf64235570c5427a1bfcc946adb1a5b201711"}}