{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:HY5BIWCUAAKNYGFTZP7UVCGO3Z","short_pith_number":"pith:HY5BIWCU","schema_version":"1.0","canonical_sha256":"3e3a1458540014dc18b3cbff4a88cede4b0a9b8ac4d71ea6fca6c642d76acb66","source":{"kind":"arxiv","id":"2507.18013","version":3},"attestation_state":"computed","paper":{"title":"Technical Report of TeleChat2, TeleChat2.5 and T1","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bingkai Yang, Chao Wang, Fubei Yao, Hanming Wu, Jiaxin Peng, Kaidong Yu, Kaipeng Jia, Mingyuan Wang, Qiyi Xie, Ruiting Song, Rui Xue, Ruiyu Fang, Shiquan Wang, Shuangyong Song, Sishi Xiong, Wenmin Deng, Xiaohui Hu, Xin Wang, Xinzhang Liu, Xuelong Li, Xuewei he, Xufeng Zhan, Yanlei Xue, Yitong Yao, Yongxiang Li, Yuhan Sun, Yuhao Jiang, Yuyao Huang, Yu Zhao, Zhaohu Xing, Zhaoxi Zhang, Zhihao Yang, Zhongjiang He, Zhuo Jiang, Zhuoru Jiang, Zhu Yuan, Zihan Wang, Zilu Huang","submitted_at":"2025-07-24T01:00:48Z","abstract_excerpt":"We introduce the latest series of TeleChat models: \\textbf{TeleChat2}, \\textbf{TeleChat2.5}, and \\textbf{T1}, offering a significant upgrade over their predecessor, TeleChat. Despite minimal changes to the model architecture, the new series achieves substantial performance gains through enhanced training strategies in both pre-training and post-training stages. The series begins with \\textbf{TeleChat2}, which undergoes pretraining on 10 trillion high-quality and diverse tokens. This is followed by Supervised Fine-Tuning (SFT) and Direct Preference Optimization (DPO) to further enhance its capa"},"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":"2507.18013","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-07-24T01:00:48Z","cross_cats_sorted":[],"title_canon_sha256":"9fb66db530671ca9aa43c31457c5d848a02ccdfcebfa88dc2f9ff5069f36b371","abstract_canon_sha256":"5001c3064726f5aae858e3413a9e907d3fb4ffbbd4dc1aff7de97ca9c8fedd67"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:44:52.573924Z","signature_b64":"/N1Z7z1fUNJaH8XE/iOf2vwT2X7Zy04fsgahdVJFBXv532cbWkKklJ2jeT0a1HF6MguqfNLTnLauVOojsVAjBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3e3a1458540014dc18b3cbff4a88cede4b0a9b8ac4d71ea6fca6c642d76acb66","last_reissued_at":"2026-07-05T11:44:52.573410Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:44:52.573410Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Technical Report of TeleChat2, TeleChat2.5 and T1","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bingkai Yang, Chao Wang, Fubei Yao, Hanming Wu, Jiaxin Peng, Kaidong Yu, Kaipeng Jia, Mingyuan Wang, Qiyi Xie, Ruiting Song, Rui Xue, Ruiyu Fang, Shiquan Wang, Shuangyong Song, Sishi Xiong, Wenmin Deng, Xiaohui Hu, Xin Wang, Xinzhang Liu, Xuelong Li, Xuewei he, Xufeng Zhan, Yanlei Xue, Yitong Yao, Yongxiang Li, Yuhan Sun, Yuhao Jiang, Yuyao Huang, Yu Zhao, Zhaohu Xing, Zhaoxi Zhang, Zhihao Yang, Zhongjiang He, Zhuo Jiang, Zhuoru Jiang, Zhu Yuan, Zihan Wang, Zilu Huang","submitted_at":"2025-07-24T01:00:48Z","abstract_excerpt":"We introduce the latest series of TeleChat models: \\textbf{TeleChat2}, \\textbf{TeleChat2.5}, and \\textbf{T1}, offering a significant upgrade over their predecessor, TeleChat. Despite minimal changes to the model architecture, the new series achieves substantial performance gains through enhanced training strategies in both pre-training and post-training stages. The series begins with \\textbf{TeleChat2}, which undergoes pretraining on 10 trillion high-quality and diverse tokens. This is followed by Supervised Fine-Tuning (SFT) and Direct Preference Optimization (DPO) to further enhance its capa"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.18013","kind":"arxiv","version":3},"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/2507.18013/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":"2507.18013","created_at":"2026-07-05T11:44:52.573475+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.18013v3","created_at":"2026-07-05T11:44:52.573475+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.18013","created_at":"2026-07-05T11:44:52.573475+00:00"},{"alias_kind":"pith_short_12","alias_value":"HY5BIWCUAAKN","created_at":"2026-07-05T11:44:52.573475+00:00"},{"alias_kind":"pith_short_16","alias_value":"HY5BIWCUAAKNYGFT","created_at":"2026-07-05T11:44:52.573475+00:00"},{"alias_kind":"pith_short_8","alias_value":"HY5BIWCU","created_at":"2026-07-05T11:44:52.573475+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2607.01392","citing_title":"Multi-Objective Exploration and Preference Optimization via Mutual Information","ref_index":121,"is_internal_anchor":false},{"citing_arxiv_id":"2606.03022","citing_title":"Hallucinations as Orthogonal Noise: Inference-Time Manifold Alignment via Dynamic Contextual Orthogonalization","ref_index":75,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01923","citing_title":"Resonant Context Anchoring: Decoupling Attention Routing and Signal Gain at Inference Time","ref_index":82,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05267","citing_title":"Bridging Generation and Training: A Systematic Review of Quality Issues in LLMs for Code","ref_index":132,"is_internal_anchor":false},{"citing_arxiv_id":"2604.14246","citing_title":"Awakening Dormant Experts:Counterfactual Routing to Mitigate MoE Hallucinations","ref_index":5,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HY5BIWCUAAKNYGFTZP7UVCGO3Z","json":"https://pith.science/pith/HY5BIWCUAAKNYGFTZP7UVCGO3Z.json","graph_json":"https://pith.science/api/pith-number/HY5BIWCUAAKNYGFTZP7UVCGO3Z/graph.json","events_json":"https://pith.science/api/pith-number/HY5BIWCUAAKNYGFTZP7UVCGO3Z/events.json","paper":"https://pith.science/paper/HY5BIWCU"},"agent_actions":{"view_html":"https://pith.science/pith/HY5BIWCUAAKNYGFTZP7UVCGO3Z","download_json":"https://pith.science/pith/HY5BIWCUAAKNYGFTZP7UVCGO3Z.json","view_paper":"https://pith.science/paper/HY5BIWCU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.18013&json=true","fetch_graph":"https://pith.science/api/pith-number/HY5BIWCUAAKNYGFTZP7UVCGO3Z/graph.json","fetch_events":"https://pith.science/api/pith-number/HY5BIWCUAAKNYGFTZP7UVCGO3Z/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HY5BIWCUAAKNYGFTZP7UVCGO3Z/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HY5BIWCUAAKNYGFTZP7UVCGO3Z/action/storage_attestation","attest_author":"https://pith.science/pith/HY5BIWCUAAKNYGFTZP7UVCGO3Z/action/author_attestation","sign_citation":"https://pith.science/pith/HY5BIWCUAAKNYGFTZP7UVCGO3Z/action/citation_signature","submit_replication":"https://pith.science/pith/HY5BIWCUAAKNYGFTZP7UVCGO3Z/action/replication_record"}},"created_at":"2026-07-05T11:44:52.573475+00:00","updated_at":"2026-07-05T11:44:52.573475+00:00"}