{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:MNAPQPW22J5ZOIGVDGIZ2TUZW4","short_pith_number":"pith:MNAPQPW2","schema_version":"1.0","canonical_sha256":"6340f83edad27b9720d519919d4e99b71b4182878028520cf7d7e913ad4ab97c","source":{"kind":"arxiv","id":"2309.11998","version":4},"attestation_state":"computed","paper":{"title":"LMSYS-Chat-1M: A Large-Scale Real-World LLM Conversation Dataset","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Eric P. Xing, Hao Zhang, Ion Stoica, Joseph E. Gonzalez, Lianmin Zheng, Siyuan Zhuang, Tianle Li, Wei-Lin Chiang, Ying Sheng, Yonghao Zhuang, Zhanghao Wu, Zhuohan Li, Zi Lin","submitted_at":"2023-09-21T12:13:55Z","abstract_excerpt":"Studying how people interact with large language models (LLMs) in real-world scenarios is increasingly important due to their widespread use in various applications. In this paper, we introduce LMSYS-Chat-1M, a large-scale dataset containing one million real-world conversations with 25 state-of-the-art LLMs. This dataset is collected from 210K unique IP addresses in the wild on our Vicuna demo and Chatbot Arena website. We offer an overview of the dataset's content, including its curation process, basic statistics, and topic distribution, highlighting its diversity, originality, and scale. We "},"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":"2309.11998","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-09-21T12:13:55Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"7a732bb685da36c597ac642bcb7721bd53947aea7e60b32b95647e705e6208b5","abstract_canon_sha256":"08bb3a2eb90cb7306379fbec7b4410d36ca5c2a92692167066d89886cae6979c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:54:03.753539Z","signature_b64":"Blxgn6KU/KmqTSLYOD3pOX2nWs4i6g1KEbJfp4Zbx3Aw2HBv2J7uBIxf5d70A/LaNaxbH/+oGPPJo8h6hNhMBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6340f83edad27b9720d519919d4e99b71b4182878028520cf7d7e913ad4ab97c","last_reissued_at":"2026-07-05T07:54:03.752979Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:54:03.752979Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LMSYS-Chat-1M: A Large-Scale Real-World LLM Conversation Dataset","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Eric P. Xing, Hao Zhang, Ion Stoica, Joseph E. Gonzalez, Lianmin Zheng, Siyuan Zhuang, Tianle Li, Wei-Lin Chiang, Ying Sheng, Yonghao Zhuang, Zhanghao Wu, Zhuohan Li, Zi Lin","submitted_at":"2023-09-21T12:13:55Z","abstract_excerpt":"Studying how people interact with large language models (LLMs) in real-world scenarios is increasingly important due to their widespread use in various applications. In this paper, we introduce LMSYS-Chat-1M, a large-scale dataset containing one million real-world conversations with 25 state-of-the-art LLMs. This dataset is collected from 210K unique IP addresses in the wild on our Vicuna demo and Chatbot Arena website. We offer an overview of the dataset's content, including its curation process, basic statistics, and topic distribution, highlighting its diversity, originality, and scale. We "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.11998","kind":"arxiv","version":4},"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/2309.11998/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":"2309.11998","created_at":"2026-07-05T07:54:03.753044+00:00"},{"alias_kind":"arxiv_version","alias_value":"2309.11998v4","created_at":"2026-07-05T07:54:03.753044+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.11998","created_at":"2026-07-05T07:54:03.753044+00:00"},{"alias_kind":"pith_short_12","alias_value":"MNAPQPW22J5Z","created_at":"2026-07-05T07:54:03.753044+00:00"},{"alias_kind":"pith_short_16","alias_value":"MNAPQPW22J5ZOIGV","created_at":"2026-07-05T07:54:03.753044+00:00"},{"alias_kind":"pith_short_8","alias_value":"MNAPQPW2","created_at":"2026-07-05T07:54:03.753044+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":29,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.26155","citing_title":"Detecting and Controlling Sycophancy with Cascading Linear Features","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2606.12247","citing_title":"Beyond Third-Person Audits: Situated Interaction Auditing for User-Centered LLM Bias Research","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2606.09659","citing_title":"End-to-End Context Compression at Scale","ref_index":95,"is_internal_anchor":false},{"citing_arxiv_id":"2606.28548","citing_title":"Turn-Averaged SAEs for Feature Discovery and Long-Context Attribution","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2605.24110","citing_title":"EvoCode-Bench: Evaluating Coding Agents in Multi-Turn Iterative Interactions","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29207","citing_title":"KernelFlume: Elastic Core-Attention Scaling for Agentic Long-Context Decoding","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2605.25247","citing_title":"Kavier: Exploring Performance, Sustainability, and Efficiency of LLM Ecosystems under Inference through Cache-Aware Discrete-Event Simulation","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2605.26788","citing_title":"SeDT: Sentence-Transformer Decision-Transformer Conditioning for Multi-Turn Conversation Reliability","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2605.28146","citing_title":"Cybersecurity AI (CAI) Dataset","ref_index":62,"is_internal_anchor":false},{"citing_arxiv_id":"2606.22748","citing_title":"AI Fiction in the Wild","ref_index":146,"is_internal_anchor":false},{"citing_arxiv_id":"2606.26396","citing_title":"At the Edge of Understanding: Sparse Autoencoders Trace The Limits of Transformer Generalization","ref_index":55,"is_internal_anchor":false},{"citing_arxiv_id":"2411.15594","citing_title":"A Survey on LLM-as-a-Judge","ref_index":221,"is_internal_anchor":false},{"citing_arxiv_id":"2504.02181","citing_title":"A Survey of Scaling in Large Language Model Reasoning","ref_index":258,"is_internal_anchor":false},{"citing_arxiv_id":"2507.18454","citing_title":"Sandwich: Joint Configuration Search and Hot-Switching for Efficient CPU LLM Serving","ref_index":69,"is_internal_anchor":false},{"citing_arxiv_id":"2601.20309","citing_title":"SuperInfer: SLO-Aware Rotary Scheduling and Memory Management for LLM Inference on Superchips","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09329","citing_title":"Test-Time Speculation","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2512.20856","citing_title":"NVIDIA Nemotron 3: Efficient and Open Intelligence","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2601.21351","citing_title":"Analytical Provisioning for Attention-FFN Disaggregated LLM Serving under Stochastic Workloads","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2603.03295","citing_title":"Language Model Goal Selection Differs from Humans' in a Self-Directed Learning Task","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2505.06120","citing_title":"LLMs Get Lost In Multi-Turn Conversation","ref_index":92,"is_internal_anchor":false},{"citing_arxiv_id":"2604.27861","citing_title":"TwinGate: Stateful Defense against Decompositional Jailbreaks in Untraceable Traffic via Asymmetric Contrastive Learning","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2604.28129","citing_title":"Latent Adversarial Detection: Adaptive Probing of LLM Activations for Multi-Turn Attack Detection","ref_index":2,"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":30,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09329","citing_title":"Test-Time Speculation","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2604.11001","citing_title":"Flow-Controlled Scheduling for LLM Inference with Provable Stability Guarantees","ref_index":30,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MNAPQPW22J5ZOIGVDGIZ2TUZW4","json":"https://pith.science/pith/MNAPQPW22J5ZOIGVDGIZ2TUZW4.json","graph_json":"https://pith.science/api/pith-number/MNAPQPW22J5ZOIGVDGIZ2TUZW4/graph.json","events_json":"https://pith.science/api/pith-number/MNAPQPW22J5ZOIGVDGIZ2TUZW4/events.json","paper":"https://pith.science/paper/MNAPQPW2"},"agent_actions":{"view_html":"https://pith.science/pith/MNAPQPW22J5ZOIGVDGIZ2TUZW4","download_json":"https://pith.science/pith/MNAPQPW22J5ZOIGVDGIZ2TUZW4.json","view_paper":"https://pith.science/paper/MNAPQPW2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2309.11998&json=true","fetch_graph":"https://pith.science/api/pith-number/MNAPQPW22J5ZOIGVDGIZ2TUZW4/graph.json","fetch_events":"https://pith.science/api/pith-number/MNAPQPW22J5ZOIGVDGIZ2TUZW4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MNAPQPW22J5ZOIGVDGIZ2TUZW4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MNAPQPW22J5ZOIGVDGIZ2TUZW4/action/storage_attestation","attest_author":"https://pith.science/pith/MNAPQPW22J5ZOIGVDGIZ2TUZW4/action/author_attestation","sign_citation":"https://pith.science/pith/MNAPQPW22J5ZOIGVDGIZ2TUZW4/action/citation_signature","submit_replication":"https://pith.science/pith/MNAPQPW22J5ZOIGVDGIZ2TUZW4/action/replication_record"}},"created_at":"2026-07-05T07:54:03.753044+00:00","updated_at":"2026-07-05T07:54:03.753044+00:00"}