{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:5A4YLTUMSN4H3QAQ3WOWFSXMFC","short_pith_number":"pith:5A4YLTUM","schema_version":"1.0","canonical_sha256":"e83985ce8c93787dc010dd9d62caec28a04c8e050f5d6ac0a213cb00316c2531","source":{"kind":"arxiv","id":"2603.29123","version":2},"attestation_state":"computed","paper":{"title":"Learning Concepts, Not Tokens: Self-Supervised Semantic Alignment for Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chen Shani, Christine Zhang, Dan Jurafsky","submitted_at":"2026-03-31T01:20:03Z","abstract_excerpt":"The next-token prediction (NTP) objective trains language models to predict a single token at each step, even though many continuations can express the same meaning. For example, in the sentence ``this sticker can be placed here'', positioned, attached, or put are all plausible alternatives. While standard NTP training treats these alternatives as mutually exclusive targets, we explore a self-supervised framework that encourages models to predict concepts, approximated as sets of semantically equivalent tokens. Models trained with this concept supervision align better with human similarity jud"},"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":"2603.29123","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-03-31T01:20:03Z","cross_cats_sorted":[],"title_canon_sha256":"9c0958d1adf10219f93675f73d40099a3445b9faed66fc12ca3241275b67eb90","abstract_canon_sha256":"a7c80835ea209eea5bf336472150761b60486fb1f532a9bd93f3f5e6e699643c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-26T02:04:09.744206Z","signature_b64":"E7+8F0PjFm0ImtWcFHNi5gmEECwkfO/DCZwP8Y0kBHpKq/uxbSc4CWU8umVKpfiZ4AZsAs8WoNjggIBytIsDBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e83985ce8c93787dc010dd9d62caec28a04c8e050f5d6ac0a213cb00316c2531","last_reissued_at":"2026-05-26T02:04:09.743429Z","signature_status":"signed_v1","first_computed_at":"2026-05-26T02:04:09.743429Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning Concepts, Not Tokens: Self-Supervised Semantic Alignment for Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chen Shani, Christine Zhang, Dan Jurafsky","submitted_at":"2026-03-31T01:20:03Z","abstract_excerpt":"The next-token prediction (NTP) objective trains language models to predict a single token at each step, even though many continuations can express the same meaning. For example, in the sentence ``this sticker can be placed here'', positioned, attached, or put are all plausible alternatives. While standard NTP training treats these alternatives as mutually exclusive targets, we explore a self-supervised framework that encourages models to predict concepts, approximated as sets of semantically equivalent tokens. Models trained with this concept supervision align better with human similarity jud"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.29123","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/2603.29123/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":"2603.29123","created_at":"2026-05-26T02:04:09.743541+00:00"},{"alias_kind":"arxiv_version","alias_value":"2603.29123v2","created_at":"2026-05-26T02:04:09.743541+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.29123","created_at":"2026-05-26T02:04:09.743541+00:00"},{"alias_kind":"pith_short_12","alias_value":"5A4YLTUMSN4H","created_at":"2026-05-26T02:04:09.743541+00:00"},{"alias_kind":"pith_short_16","alias_value":"5A4YLTUMSN4H3QAQ","created_at":"2026-05-26T02:04:09.743541+00:00"},{"alias_kind":"pith_short_8","alias_value":"5A4YLTUM","created_at":"2026-05-26T02:04:09.743541+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/5A4YLTUMSN4H3QAQ3WOWFSXMFC","json":"https://pith.science/pith/5A4YLTUMSN4H3QAQ3WOWFSXMFC.json","graph_json":"https://pith.science/api/pith-number/5A4YLTUMSN4H3QAQ3WOWFSXMFC/graph.json","events_json":"https://pith.science/api/pith-number/5A4YLTUMSN4H3QAQ3WOWFSXMFC/events.json","paper":"https://pith.science/paper/5A4YLTUM"},"agent_actions":{"view_html":"https://pith.science/pith/5A4YLTUMSN4H3QAQ3WOWFSXMFC","download_json":"https://pith.science/pith/5A4YLTUMSN4H3QAQ3WOWFSXMFC.json","view_paper":"https://pith.science/paper/5A4YLTUM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2603.29123&json=true","fetch_graph":"https://pith.science/api/pith-number/5A4YLTUMSN4H3QAQ3WOWFSXMFC/graph.json","fetch_events":"https://pith.science/api/pith-number/5A4YLTUMSN4H3QAQ3WOWFSXMFC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5A4YLTUMSN4H3QAQ3WOWFSXMFC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5A4YLTUMSN4H3QAQ3WOWFSXMFC/action/storage_attestation","attest_author":"https://pith.science/pith/5A4YLTUMSN4H3QAQ3WOWFSXMFC/action/author_attestation","sign_citation":"https://pith.science/pith/5A4YLTUMSN4H3QAQ3WOWFSXMFC/action/citation_signature","submit_replication":"https://pith.science/pith/5A4YLTUMSN4H3QAQ3WOWFSXMFC/action/replication_record"}},"created_at":"2026-05-26T02:04:09.743541+00:00","updated_at":"2026-05-26T02:04:09.743541+00:00"}