{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:VGEWA5ZYQHAPL67SX7YV3ZZ3HF","short_pith_number":"pith:VGEWA5ZY","schema_version":"1.0","canonical_sha256":"a98960773881c0f5fbf2bff15de73b394e0e4790a9a06a14c93aac0132806345","source":{"kind":"arxiv","id":"2507.07432","version":2},"attestation_state":"computed","paper":{"title":"Neural networks leverage nominally quantum and post-quantum representations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["quant-ph"],"primary_cat":"cs.LG","authors_text":"Adam S. Shai, Paul M. Riechers, Thomas J. Elliott","submitted_at":"2025-07-10T05:09:19Z","abstract_excerpt":"We show that deep neural networks, including transformers and RNNs, pretrained as usual on next-token prediction, intrinsically discover and represent beliefs over 'quantum' and 'post-quantum' low-dimensional generative models of their training data -- as if performing iterative Bayesian updates over the latent state of this world model during inference as they observe more context. Notably, neural nets easily find these representation whereas there is no finite classical circuit that would do the job. The corresponding geometric relationships among neural activations induced by different inpu"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2507.07432","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-07-10T05:09:19Z","cross_cats_sorted":["quant-ph"],"title_canon_sha256":"645d5d86e2d1e7ea950147905f1e5e6aa0905317edcb4f7f424ea17354085dc2","abstract_canon_sha256":"4c241cb7ab230c3ec4f5f20808ccec355f259bad0d808e7e8569f2ff1f020a85"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:35:58.231498Z","signature_b64":"nOah2Q7aGzDba0zwum6KjzNWoICyp5Ve4bdwHM/8zkvf1wG2ggIjGxoudI+iRVk3F1y1vvZBz+/v5BGYYA49Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a98960773881c0f5fbf2bff15de73b394e0e4790a9a06a14c93aac0132806345","last_reissued_at":"2026-07-05T11:35:58.231078Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:35:58.231078Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Neural networks leverage nominally quantum and post-quantum representations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["quant-ph"],"primary_cat":"cs.LG","authors_text":"Adam S. Shai, Paul M. Riechers, Thomas J. Elliott","submitted_at":"2025-07-10T05:09:19Z","abstract_excerpt":"We show that deep neural networks, including transformers and RNNs, pretrained as usual on next-token prediction, intrinsically discover and represent beliefs over 'quantum' and 'post-quantum' low-dimensional generative models of their training data -- as if performing iterative Bayesian updates over the latent state of this world model during inference as they observe more context. Notably, neural nets easily find these representation whereas there is no finite classical circuit that would do the job. The corresponding geometric relationships among neural activations induced by different inpu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.07432","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/2507.07432/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2507.07432","created_at":"2026-07-05T11:35:58.231134+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.07432v2","created_at":"2026-07-05T11:35:58.231134+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.07432","created_at":"2026-07-05T11:35:58.231134+00:00"},{"alias_kind":"pith_short_12","alias_value":"VGEWA5ZYQHAP","created_at":"2026-07-05T11:35:58.231134+00:00"},{"alias_kind":"pith_short_16","alias_value":"VGEWA5ZYQHAPL67S","created_at":"2026-07-05T11:35:58.231134+00:00"},{"alias_kind":"pith_short_8","alias_value":"VGEWA5ZY","created_at":"2026-07-05T11:35:58.231134+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.01302","citing_title":"Structure and Scale in Simplicial Sequence Modelling","ref_index":56,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VGEWA5ZYQHAPL67SX7YV3ZZ3HF","json":"https://pith.science/pith/VGEWA5ZYQHAPL67SX7YV3ZZ3HF.json","graph_json":"https://pith.science/api/pith-number/VGEWA5ZYQHAPL67SX7YV3ZZ3HF/graph.json","events_json":"https://pith.science/api/pith-number/VGEWA5ZYQHAPL67SX7YV3ZZ3HF/events.json","paper":"https://pith.science/paper/VGEWA5ZY"},"agent_actions":{"view_html":"https://pith.science/pith/VGEWA5ZYQHAPL67SX7YV3ZZ3HF","download_json":"https://pith.science/pith/VGEWA5ZYQHAPL67SX7YV3ZZ3HF.json","view_paper":"https://pith.science/paper/VGEWA5ZY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.07432&json=true","fetch_graph":"https://pith.science/api/pith-number/VGEWA5ZYQHAPL67SX7YV3ZZ3HF/graph.json","fetch_events":"https://pith.science/api/pith-number/VGEWA5ZYQHAPL67SX7YV3ZZ3HF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VGEWA5ZYQHAPL67SX7YV3ZZ3HF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VGEWA5ZYQHAPL67SX7YV3ZZ3HF/action/storage_attestation","attest_author":"https://pith.science/pith/VGEWA5ZYQHAPL67SX7YV3ZZ3HF/action/author_attestation","sign_citation":"https://pith.science/pith/VGEWA5ZYQHAPL67SX7YV3ZZ3HF/action/citation_signature","submit_replication":"https://pith.science/pith/VGEWA5ZYQHAPL67SX7YV3ZZ3HF/action/replication_record"}},"created_at":"2026-07-05T11:35:58.231134+00:00","updated_at":"2026-07-05T11:35:58.231134+00:00"}