{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:NCB355APP2FBL6EXUJGNO5C75D","short_pith_number":"pith:NCB355AP","canonical_record":{"source":{"id":"2605.19394","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-19T05:40:12Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"ee34a0714ae02233056bf3f0dd0356d8d9e7ad6303ae81694c8115ba10530a60","abstract_canon_sha256":"e86b514e222667e41bb177c83c88f69307cd77e6f7b7568eb85a90000500ad74"},"schema_version":"1.0"},"canonical_sha256":"6883bef40f7e8a15f897a24cd7745fe8dc124f5181db4dd1178f1bac5e3f5d01","source":{"kind":"arxiv","id":"2605.19394","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.19394","created_at":"2026-05-20T01:05:43Z"},{"alias_kind":"arxiv_version","alias_value":"2605.19394v1","created_at":"2026-05-20T01:05:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.19394","created_at":"2026-05-20T01:05:43Z"},{"alias_kind":"pith_short_12","alias_value":"NCB355APP2FB","created_at":"2026-05-20T01:05:43Z"},{"alias_kind":"pith_short_16","alias_value":"NCB355APP2FBL6EX","created_at":"2026-05-20T01:05:43Z"},{"alias_kind":"pith_short_8","alias_value":"NCB355AP","created_at":"2026-05-20T01:05:43Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:NCB355APP2FBL6EXUJGNO5C75D","target":"record","payload":{"canonical_record":{"source":{"id":"2605.19394","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-19T05:40:12Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"ee34a0714ae02233056bf3f0dd0356d8d9e7ad6303ae81694c8115ba10530a60","abstract_canon_sha256":"e86b514e222667e41bb177c83c88f69307cd77e6f7b7568eb85a90000500ad74"},"schema_version":"1.0"},"canonical_sha256":"6883bef40f7e8a15f897a24cd7745fe8dc124f5181db4dd1178f1bac5e3f5d01","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T01:05:43.496688Z","signature_b64":"iS1sRVTbxrfSydfgnvTTB+yqPgZbENnEirTq6TbknBuUVcZ5VLgyRMLovLQciBKOM/v3wyD5A8LWV/53axofDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6883bef40f7e8a15f897a24cd7745fe8dc124f5181db4dd1178f1bac5e3f5d01","last_reissued_at":"2026-05-20T01:05:43.495825Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T01:05:43.495825Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.19394","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":"2026-05-20T01:05:43Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PqIXg0s1ek8gnTQUW4zjE4GMJE0NjRnQ8yJVb3/UjaQ9npBNLYrWH6VAmoTGArW0I3T9CHDBXd57lZ/rFmIbCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-21T09:02:15.624055Z"},"content_sha256":"5d3a29ce2427671bf90effc70ffe5edfa28ab316bbfb8b42d956a1fdebab0d04","schema_version":"1.0","event_id":"sha256:5d3a29ce2427671bf90effc70ffe5edfa28ab316bbfb8b42d956a1fdebab0d04"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:NCB355APP2FBL6EXUJGNO5C75D","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"EmbGen: Teaching with Reassembled Corpora","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Andrea Nicastro, Anna Leontjeva, Arun K Lenin, Kai Rouse","submitted_at":"2026-05-19T05:40:12Z","abstract_excerpt":"Adapting small instruction-tuned models to specialized domains often relies on supervised fine-tuning (SFT) on curated instruction-response examples, which is expensive to collect at scale. Synthetic training examples generated by a teacher LLM from a domain corpus can reduce this cost, but existing pipelines can produce homogenized outputs and do not consistently capture cross-passage or cross-document dependencies. We introduce EmbGen, a synthetic data generation pipeline that decomposes a corpus into entity-description pairs, reassembles them using semantic structure inferred from embedding"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.19394","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/2605.19394/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":"2026-05-20T01:05:43Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"FdUVlJC0lHIfh5fRDbCWq4emS2m5YroVfHj1Hc8hSd2Ig16gIwejDbYIGjyeuIBqIhZPeEE4VhoAQbI5OWiOBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-21T09:02:15.624479Z"},"content_sha256":"bbcfe2b586c33deffd3c191958bf1e92d77dd9a70708939167bccf71b8d8f270","schema_version":"1.0","event_id":"sha256:bbcfe2b586c33deffd3c191958bf1e92d77dd9a70708939167bccf71b8d8f270"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/NCB355APP2FBL6EXUJGNO5C75D/bundle.json","state_url":"https://pith.science/pith/NCB355APP2FBL6EXUJGNO5C75D/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/NCB355APP2FBL6EXUJGNO5C75D/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":"2026-05-21T09:02:15Z","links":{"resolver":"https://pith.science/pith/NCB355APP2FBL6EXUJGNO5C75D","bundle":"https://pith.science/pith/NCB355APP2FBL6EXUJGNO5C75D/bundle.json","state":"https://pith.science/pith/NCB355APP2FBL6EXUJGNO5C75D/state.json","well_known_bundle":"https://pith.science/.well-known/pith/NCB355APP2FBL6EXUJGNO5C75D/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:NCB355APP2FBL6EXUJGNO5C75D","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":"e86b514e222667e41bb177c83c88f69307cd77e6f7b7568eb85a90000500ad74","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-19T05:40:12Z","title_canon_sha256":"ee34a0714ae02233056bf3f0dd0356d8d9e7ad6303ae81694c8115ba10530a60"},"schema_version":"1.0","source":{"id":"2605.19394","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.19394","created_at":"2026-05-20T01:05:43Z"},{"alias_kind":"arxiv_version","alias_value":"2605.19394v1","created_at":"2026-05-20T01:05:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.19394","created_at":"2026-05-20T01:05:43Z"},{"alias_kind":"pith_short_12","alias_value":"NCB355APP2FB","created_at":"2026-05-20T01:05:43Z"},{"alias_kind":"pith_short_16","alias_value":"NCB355APP2FBL6EX","created_at":"2026-05-20T01:05:43Z"},{"alias_kind":"pith_short_8","alias_value":"NCB355AP","created_at":"2026-05-20T01:05:43Z"}],"graph_snapshots":[{"event_id":"sha256:bbcfe2b586c33deffd3c191958bf1e92d77dd9a70708939167bccf71b8d8f270","target":"graph","created_at":"2026-05-20T01:05:43Z","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/2605.19394/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Adapting small instruction-tuned models to specialized domains often relies on supervised fine-tuning (SFT) on curated instruction-response examples, which is expensive to collect at scale. Synthetic training examples generated by a teacher LLM from a domain corpus can reduce this cost, but existing pipelines can produce homogenized outputs and do not consistently capture cross-passage or cross-document dependencies. We introduce EmbGen, a synthetic data generation pipeline that decomposes a corpus into entity-description pairs, reassembles them using semantic structure inferred from embedding","authors_text":"Andrea Nicastro, Anna Leontjeva, Arun K Lenin, Kai Rouse","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-19T05:40:12Z","title":"EmbGen: Teaching with Reassembled Corpora"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.19394","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":"sha256:5d3a29ce2427671bf90effc70ffe5edfa28ab316bbfb8b42d956a1fdebab0d04","target":"record","created_at":"2026-05-20T01:05:43Z","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":"e86b514e222667e41bb177c83c88f69307cd77e6f7b7568eb85a90000500ad74","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-19T05:40:12Z","title_canon_sha256":"ee34a0714ae02233056bf3f0dd0356d8d9e7ad6303ae81694c8115ba10530a60"},"schema_version":"1.0","source":{"id":"2605.19394","kind":"arxiv","version":1}},"canonical_sha256":"6883bef40f7e8a15f897a24cd7745fe8dc124f5181db4dd1178f1bac5e3f5d01","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6883bef40f7e8a15f897a24cd7745fe8dc124f5181db4dd1178f1bac5e3f5d01","first_computed_at":"2026-05-20T01:05:43.495825Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T01:05:43.495825Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"iS1sRVTbxrfSydfgnvTTB+yqPgZbENnEirTq6TbknBuUVcZ5VLgyRMLovLQciBKOM/v3wyD5A8LWV/53axofDQ==","signature_status":"signed_v1","signed_at":"2026-05-20T01:05:43.496688Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.19394","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5d3a29ce2427671bf90effc70ffe5edfa28ab316bbfb8b42d956a1fdebab0d04","sha256:bbcfe2b586c33deffd3c191958bf1e92d77dd9a70708939167bccf71b8d8f270"],"state_sha256":"98b7608b7185566f8e251b77213353ced7173ac481bd386dc6e8e2e5fd02758a"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"W1JFwk5IDbyWpNw8CFp6XS3rSBInbBOt9SBLLsTPR8cUM3/eTi4l9vVROkBwzxWXa3yDJsdao5xavNPOLpebCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-21T09:02:15.626681Z","bundle_sha256":"069f011000f29e2c370565a38dc896b60eeb763e6ae738887fd6cbdd869e6db2"}}