{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:XYKQSRSCHN35A7MRIDKQ2IFRMS","short_pith_number":"pith:XYKQSRSC","schema_version":"1.0","canonical_sha256":"be150946423b77d07d9140d50d20b1649764bd8c7fa2f2d3973bcce90f899286","source":{"kind":"arxiv","id":"2606.09731","version":1},"attestation_state":"computed","paper":{"title":"Tight Sample Complexity of Transformers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Chenxiao Yang, Nathan Srebro, Zhiyuan Li","submitted_at":"2026-06-08T16:56:24Z","abstract_excerpt":"We tightly characterize the VC dimension of depth-$L$ Transformers with a total of $W$ parameters, mapping an input sequence of length $T$ to a single output, establishing an upper bound of $O(L W \\log (T W))$ and a nearly matching lower bound of $\\Omega(L W \\log (T W / L))$. We further tightly characterize the sample complexity of chain-of-thought learning using such a Transformer, showing teacher forcing (i.e. selecting a predictor consistent with the entire chain-of-thought on training data) learns with sample complexity $O\\left(L W \\log \\left(\\left(T+T^{\\prime}\\right) W\\right)\\right)$ and "},"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":"2606.09731","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-06-08T16:56:24Z","cross_cats_sorted":[],"title_canon_sha256":"5674a29ef33c8576b13ab8c21c6fd8b7e03bd4d2a668fb041b43b5aa9b12f803","abstract_canon_sha256":"cad8aa3efad03b92d44dd1cbd591cb114ea43df7f9a576399f7826153f65948d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-09T02:09:06.514546Z","signature_b64":"DhH0OcUtPq8R5M7YsUwSYDMz/eUslzacfGmypVvtXHR7UfP4RCXtQjhZ8OVQQRWlddZfdDBWUeg+/lps3LpsDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"be150946423b77d07d9140d50d20b1649764bd8c7fa2f2d3973bcce90f899286","last_reissued_at":"2026-06-09T02:09:06.513607Z","signature_status":"signed_v1","first_computed_at":"2026-06-09T02:09:06.513607Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Tight Sample Complexity of Transformers","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Chenxiao Yang, Nathan Srebro, Zhiyuan Li","submitted_at":"2026-06-08T16:56:24Z","abstract_excerpt":"We tightly characterize the VC dimension of depth-$L$ Transformers with a total of $W$ parameters, mapping an input sequence of length $T$ to a single output, establishing an upper bound of $O(L W \\log (T W))$ and a nearly matching lower bound of $\\Omega(L W \\log (T W / L))$. We further tightly characterize the sample complexity of chain-of-thought learning using such a Transformer, showing teacher forcing (i.e. selecting a predictor consistent with the entire chain-of-thought on training data) learns with sample complexity $O\\left(L W \\log \\left(\\left(T+T^{\\prime}\\right) W\\right)\\right)$ and "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.09731","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/2606.09731/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":"2606.09731","created_at":"2026-06-09T02:09:06.513764+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.09731v1","created_at":"2026-06-09T02:09:06.513764+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.09731","created_at":"2026-06-09T02:09:06.513764+00:00"},{"alias_kind":"pith_short_12","alias_value":"XYKQSRSCHN35","created_at":"2026-06-09T02:09:06.513764+00:00"},{"alias_kind":"pith_short_16","alias_value":"XYKQSRSCHN35A7MR","created_at":"2026-06-09T02:09:06.513764+00:00"},{"alias_kind":"pith_short_8","alias_value":"XYKQSRSC","created_at":"2026-06-09T02:09:06.513764+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XYKQSRSCHN35A7MRIDKQ2IFRMS","json":"https://pith.science/pith/XYKQSRSCHN35A7MRIDKQ2IFRMS.json","graph_json":"https://pith.science/api/pith-number/XYKQSRSCHN35A7MRIDKQ2IFRMS/graph.json","events_json":"https://pith.science/api/pith-number/XYKQSRSCHN35A7MRIDKQ2IFRMS/events.json","paper":"https://pith.science/paper/XYKQSRSC"},"agent_actions":{"view_html":"https://pith.science/pith/XYKQSRSCHN35A7MRIDKQ2IFRMS","download_json":"https://pith.science/pith/XYKQSRSCHN35A7MRIDKQ2IFRMS.json","view_paper":"https://pith.science/paper/XYKQSRSC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.09731&json=true","fetch_graph":"https://pith.science/api/pith-number/XYKQSRSCHN35A7MRIDKQ2IFRMS/graph.json","fetch_events":"https://pith.science/api/pith-number/XYKQSRSCHN35A7MRIDKQ2IFRMS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XYKQSRSCHN35A7MRIDKQ2IFRMS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XYKQSRSCHN35A7MRIDKQ2IFRMS/action/storage_attestation","attest_author":"https://pith.science/pith/XYKQSRSCHN35A7MRIDKQ2IFRMS/action/author_attestation","sign_citation":"https://pith.science/pith/XYKQSRSCHN35A7MRIDKQ2IFRMS/action/citation_signature","submit_replication":"https://pith.science/pith/XYKQSRSCHN35A7MRIDKQ2IFRMS/action/replication_record"}},"created_at":"2026-06-09T02:09:06.513764+00:00","updated_at":"2026-06-09T02:09:06.513764+00:00"}