{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:KXCA3YX4QTPNWPNAHBEVD5Y2OA","short_pith_number":"pith:KXCA3YX4","schema_version":"1.0","canonical_sha256":"55c40de2fc84dedb3da0384951f71a703cff6c986bdf52792aa0532b5f949201","source":{"kind":"arxiv","id":"2606.03080","version":1},"attestation_state":"computed","paper":{"title":"Regret Pre-training: Bridging Prior and Posterior Views for Enhanced Knowledge Grounding","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Jiaxuan Li, Jiayin Wang, Mingkuan Zhao, Suquan Chen, Wentao Hu, Xiaoyan Zhu, Xiayu Sun, Xin Lai","submitted_at":"2026-06-02T03:11:39Z","abstract_excerpt":"Causal language models factorize sequence probabilities using only preceding context, leaving future information unexploited during training despite its availability in the training data. This paper introduces Regret Pre-training, a self-supervised framework grounded in the Learning Using Privileged Information (LUPI) paradigm. The framework employs a dual-view architecture in which a single model generates both a causal Student distribution and a future-conditioned Teacher distribution. The training objective augments standard language modeling with a regret loss that minimizes the KL diverge"},"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":"2606.03080","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-02T03:11:39Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b2751ff0856beb2cdaf9ba1da88bf8bddd59e5f798520a6530a489009906de53","abstract_canon_sha256":"c818ab427159a4063fd5b6e1bc43f6a5afc0f90a5203462ab3b627b8ffbca66e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-03T01:05:31.105852Z","signature_b64":"J1PHHbw2U5IbtY8UY5n9hEwl9nPYu0F0oKl56YwwQaJ+3h4Y/kWZlgpjwusAnY02DEfuuPWo8Olifo6mdyfcAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"55c40de2fc84dedb3da0384951f71a703cff6c986bdf52792aa0532b5f949201","last_reissued_at":"2026-06-03T01:05:31.105440Z","signature_status":"signed_v1","first_computed_at":"2026-06-03T01:05:31.105440Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Regret Pre-training: Bridging Prior and Posterior Views for Enhanced Knowledge Grounding","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Jiaxuan Li, Jiayin Wang, Mingkuan Zhao, Suquan Chen, Wentao Hu, Xiaoyan Zhu, Xiayu Sun, Xin Lai","submitted_at":"2026-06-02T03:11:39Z","abstract_excerpt":"Causal language models factorize sequence probabilities using only preceding context, leaving future information unexploited during training despite its availability in the training data. This paper introduces Regret Pre-training, a self-supervised framework grounded in the Learning Using Privileged Information (LUPI) paradigm. The framework employs a dual-view architecture in which a single model generates both a causal Student distribution and a future-conditioned Teacher distribution. The training objective augments standard language modeling with a regret loss that minimizes the KL diverge"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.03080","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/2606.03080/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":"2606.03080","created_at":"2026-06-03T01:05:31.105492+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.03080v1","created_at":"2026-06-03T01:05:31.105492+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.03080","created_at":"2026-06-03T01:05:31.105492+00:00"},{"alias_kind":"pith_short_12","alias_value":"KXCA3YX4QTPN","created_at":"2026-06-03T01:05:31.105492+00:00"},{"alias_kind":"pith_short_16","alias_value":"KXCA3YX4QTPNWPNA","created_at":"2026-06-03T01:05:31.105492+00:00"},{"alias_kind":"pith_short_8","alias_value":"KXCA3YX4","created_at":"2026-06-03T01:05:31.105492+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/KXCA3YX4QTPNWPNAHBEVD5Y2OA","json":"https://pith.science/pith/KXCA3YX4QTPNWPNAHBEVD5Y2OA.json","graph_json":"https://pith.science/api/pith-number/KXCA3YX4QTPNWPNAHBEVD5Y2OA/graph.json","events_json":"https://pith.science/api/pith-number/KXCA3YX4QTPNWPNAHBEVD5Y2OA/events.json","paper":"https://pith.science/paper/KXCA3YX4"},"agent_actions":{"view_html":"https://pith.science/pith/KXCA3YX4QTPNWPNAHBEVD5Y2OA","download_json":"https://pith.science/pith/KXCA3YX4QTPNWPNAHBEVD5Y2OA.json","view_paper":"https://pith.science/paper/KXCA3YX4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.03080&json=true","fetch_graph":"https://pith.science/api/pith-number/KXCA3YX4QTPNWPNAHBEVD5Y2OA/graph.json","fetch_events":"https://pith.science/api/pith-number/KXCA3YX4QTPNWPNAHBEVD5Y2OA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KXCA3YX4QTPNWPNAHBEVD5Y2OA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KXCA3YX4QTPNWPNAHBEVD5Y2OA/action/storage_attestation","attest_author":"https://pith.science/pith/KXCA3YX4QTPNWPNAHBEVD5Y2OA/action/author_attestation","sign_citation":"https://pith.science/pith/KXCA3YX4QTPNWPNAHBEVD5Y2OA/action/citation_signature","submit_replication":"https://pith.science/pith/KXCA3YX4QTPNWPNAHBEVD5Y2OA/action/replication_record"}},"created_at":"2026-06-03T01:05:31.105492+00:00","updated_at":"2026-06-03T01:05:31.105492+00:00"}