{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:VM2FN2YZPZ7FBEHQZ2DFI5Z3NH","short_pith_number":"pith:VM2FN2YZ","schema_version":"1.0","canonical_sha256":"ab3456eb197e7e5090f0ce8654773b69f4656eaa9a87b1db070cfd2ad08c50b1","source":{"kind":"arxiv","id":"2506.11116","version":1},"attestation_state":"computed","paper":{"title":"Infinity Instruct: Scaling Instruction Selection and Synthesis to Enhance Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Bo-Wen Zhang, Boyan Gao, Guang Liu, Hanyu Zhao, Jijie Li, Liangdong Wang, Li Du, Yonghua Lin","submitted_at":"2025-06-09T06:37:15Z","abstract_excerpt":"Large Language Models (LLMs) demonstrate strong performance in real-world applications, yet existing open-source instruction datasets often concentrate on narrow domains, such as mathematics or coding, limiting generalization and widening the gap with proprietary models. To bridge this gap, we introduce Infinity-Instruct, a high-quality instruction dataset designed to enhance both foundational and chat capabilities of LLMs through a two-phase pipeline. In Phase 1, we curate 7.4M high-quality foundational instructions (InfInstruct-F-7.4M) from over 100M samples using hybrid data selection techn"},"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":"2506.11116","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-09T06:37:15Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"d00ac5d7961b58d2f7cd428ad2f04a802737e7f8bc8e28f91042ce271cf328c4","abstract_canon_sha256":"af84c1a9fd0f330070e3853050400f4a7665721221e2af2871f1d788e2dbd795"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:20:59.726267Z","signature_b64":"Vng+ZNMDLWW8KqYyqMwhvaUwJ2SKH9EmYdErWuM/Bn3hrT7fa5/5+OCmmaALh7rMa5TJNbVjjCei4l/UsmmpDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ab3456eb197e7e5090f0ce8654773b69f4656eaa9a87b1db070cfd2ad08c50b1","last_reissued_at":"2026-07-05T11:20:59.725514Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:20:59.725514Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Infinity Instruct: Scaling Instruction Selection and Synthesis to Enhance Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Bo-Wen Zhang, Boyan Gao, Guang Liu, Hanyu Zhao, Jijie Li, Liangdong Wang, Li Du, Yonghua Lin","submitted_at":"2025-06-09T06:37:15Z","abstract_excerpt":"Large Language Models (LLMs) demonstrate strong performance in real-world applications, yet existing open-source instruction datasets often concentrate on narrow domains, such as mathematics or coding, limiting generalization and widening the gap with proprietary models. To bridge this gap, we introduce Infinity-Instruct, a high-quality instruction dataset designed to enhance both foundational and chat capabilities of LLMs through a two-phase pipeline. In Phase 1, we curate 7.4M high-quality foundational instructions (InfInstruct-F-7.4M) from over 100M samples using hybrid data selection techn"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.11116","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/2506.11116/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":"2506.11116","created_at":"2026-07-05T11:20:59.725577+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.11116v1","created_at":"2026-07-05T11:20:59.725577+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.11116","created_at":"2026-07-05T11:20:59.725577+00:00"},{"alias_kind":"pith_short_12","alias_value":"VM2FN2YZPZ7F","created_at":"2026-07-05T11:20:59.725577+00:00"},{"alias_kind":"pith_short_16","alias_value":"VM2FN2YZPZ7FBEHQ","created_at":"2026-07-05T11:20:59.725577+00:00"},{"alias_kind":"pith_short_8","alias_value":"VM2FN2YZ","created_at":"2026-07-05T11:20:59.725577+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":12,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.07531","citing_title":"mllm-shap: A Shapley Value Explainability Platform for Text-Audio Multimodal Large Language Models","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2607.02118","citing_title":"Enhancing Fitness Intelligence through Domain-Specific LLM Post-Training","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14531","citing_title":"Language Generation as Optimal Control: Closed-Loop Diffusion in Latent Control Space","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04062","citing_title":"EdgeRazor: A Lightweight Framework for Large Language Models via Mixed-Precision Quantization-Aware Distillation","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2512.17843","citing_title":"ShareChat: A Dataset of Chatbot Conversations in the Wild","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14531","citing_title":"Language Generation as Optimal Control: Closed-Loop Diffusion in Latent Control Space","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2601.21503","citing_title":"MAR: Efficient Large Language Models via Module-aware Architecture Refinement","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14531","citing_title":"Language Generation as Optimal Control: Closed-Loop Diffusion in Latent Control Space","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09635","citing_title":"K12-KGraph: A Curriculum-Aligned Knowledge Graph for Benchmarking and Training Educational LLMs","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10453","citing_title":"SlimSpec: Low-Rank Draft LM-Head for Accelerated Speculative Decoding","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04062","citing_title":"EdgeRazor: A Lightweight Framework for Large Language Models via Mixed-Precision Quantization-Aware Distillation","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06765","citing_title":"VITA-QinYu: Expressive Spoken Language Model for Role-Playing and Singing","ref_index":111,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VM2FN2YZPZ7FBEHQZ2DFI5Z3NH","json":"https://pith.science/pith/VM2FN2YZPZ7FBEHQZ2DFI5Z3NH.json","graph_json":"https://pith.science/api/pith-number/VM2FN2YZPZ7FBEHQZ2DFI5Z3NH/graph.json","events_json":"https://pith.science/api/pith-number/VM2FN2YZPZ7FBEHQZ2DFI5Z3NH/events.json","paper":"https://pith.science/paper/VM2FN2YZ"},"agent_actions":{"view_html":"https://pith.science/pith/VM2FN2YZPZ7FBEHQZ2DFI5Z3NH","download_json":"https://pith.science/pith/VM2FN2YZPZ7FBEHQZ2DFI5Z3NH.json","view_paper":"https://pith.science/paper/VM2FN2YZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.11116&json=true","fetch_graph":"https://pith.science/api/pith-number/VM2FN2YZPZ7FBEHQZ2DFI5Z3NH/graph.json","fetch_events":"https://pith.science/api/pith-number/VM2FN2YZPZ7FBEHQZ2DFI5Z3NH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VM2FN2YZPZ7FBEHQZ2DFI5Z3NH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VM2FN2YZPZ7FBEHQZ2DFI5Z3NH/action/storage_attestation","attest_author":"https://pith.science/pith/VM2FN2YZPZ7FBEHQZ2DFI5Z3NH/action/author_attestation","sign_citation":"https://pith.science/pith/VM2FN2YZPZ7FBEHQZ2DFI5Z3NH/action/citation_signature","submit_replication":"https://pith.science/pith/VM2FN2YZPZ7FBEHQZ2DFI5Z3NH/action/replication_record"}},"created_at":"2026-07-05T11:20:59.725577+00:00","updated_at":"2026-07-05T11:20:59.725577+00:00"}