{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:YOSF547IIPN4WTGWEVOA2WM777","short_pith_number":"pith:YOSF547I","schema_version":"1.0","canonical_sha256":"c3a45ef3e843dbcb4cd6255c0d599fffe5111c4867ff091b745e516b4ddc9c15","source":{"kind":"arxiv","id":"2501.17399","version":2},"attestation_state":"computed","paper":{"title":"MultiChallenge: A Realistic Multi-Turn Conversation Evaluation Benchmark Challenging to Frontier LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chen Xing, Dean Lee, Ed-Yeremai Cardona, Jeremy Kritz, Johannes Mols, Kaustubh Deshpande, Lifeng Jin, Summer Yue, Ved Sirdeshmukh, Willow Primack","submitted_at":"2025-01-29T03:29:24Z","abstract_excerpt":"We present MultiChallenge, a pioneering benchmark evaluating large language models (LLMs) on conducting multi-turn conversations with human users, a crucial yet underexamined capability for their applications. MultiChallenge identifies four categories of challenges in multi-turn conversations that are not only common and realistic among current human-LLM interactions, but are also challenging to all current frontier LLMs. All 4 challenges require accurate instruction-following, context allocation, and in-context reasoning at the same time. We also develop LLM as judge with instance-level rubri"},"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":"2501.17399","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-01-29T03:29:24Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"1876ea3cdadd367a70f20ae235f009299e38c7976602a79d4fe8ef50ece8d788","abstract_canon_sha256":"693cc7bd822fa027a0216bbc27f8e2cad30cf2b70cdc18288dad73810d5a94a7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:25:16.631767Z","signature_b64":"XcchUoCWqf8WCcN3EO8R4g06Mq4rTWwvtfzEtqlNDIwgWfsltnFTKi227o03weOIDv0apbtwlaZ1GNyrpsyiBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c3a45ef3e843dbcb4cd6255c0d599fffe5111c4867ff091b745e516b4ddc9c15","last_reissued_at":"2026-07-05T10:25:16.631098Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:25:16.631098Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MultiChallenge: A Realistic Multi-Turn Conversation Evaluation Benchmark Challenging to Frontier LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chen Xing, Dean Lee, Ed-Yeremai Cardona, Jeremy Kritz, Johannes Mols, Kaustubh Deshpande, Lifeng Jin, Summer Yue, Ved Sirdeshmukh, Willow Primack","submitted_at":"2025-01-29T03:29:24Z","abstract_excerpt":"We present MultiChallenge, a pioneering benchmark evaluating large language models (LLMs) on conducting multi-turn conversations with human users, a crucial yet underexamined capability for their applications. MultiChallenge identifies four categories of challenges in multi-turn conversations that are not only common and realistic among current human-LLM interactions, but are also challenging to all current frontier LLMs. All 4 challenges require accurate instruction-following, context allocation, and in-context reasoning at the same time. We also develop LLM as judge with instance-level rubri"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.17399","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/2501.17399/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":"2501.17399","created_at":"2026-07-05T10:25:16.631172+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.17399v2","created_at":"2026-07-05T10:25:16.631172+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.17399","created_at":"2026-07-05T10:25:16.631172+00:00"},{"alias_kind":"pith_short_12","alias_value":"YOSF547IIPN4","created_at":"2026-07-05T10:25:16.631172+00:00"},{"alias_kind":"pith_short_16","alias_value":"YOSF547IIPN4WTGW","created_at":"2026-07-05T10:25:16.631172+00:00"},{"alias_kind":"pith_short_8","alias_value":"YOSF547I","created_at":"2026-07-05T10:25:16.631172+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":11,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.18216","citing_title":"Zone of Proximal Policy Optimization: Teacher in Prompts, Not Gradients","ref_index":155,"is_internal_anchor":false},{"citing_arxiv_id":"2606.17114","citing_title":"An Evaluation of Data Leakage Risks in Tool-Using LLM Agents in Realistic Scenarios","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2606.09878","citing_title":"FailureScope: Cross-Regime Behavioral Diagnosis of Language Model Weaknesses","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2605.27671","citing_title":"Evolving and Detecting Multi-Turn Deception using Geometric Signatures","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2604.16310","citing_title":"RAG-DIVE: A Dynamic Approach for Multi-Turn Dialogue Evaluation in Retrieval-Augmented Generation","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2511.20857","citing_title":"Evo-Memory: Benchmarking LLM Agent Test-time Learning with Self-Evolving Memory","ref_index":131,"is_internal_anchor":false},{"citing_arxiv_id":"2505.06120","citing_title":"LLMs Get Lost In Multi-Turn Conversation","ref_index":76,"is_internal_anchor":false},{"citing_arxiv_id":"2507.17746","citing_title":"Rubrics as Rewards: Reinforcement Learning Beyond Verifiable Domains","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2505.20275","citing_title":"ImgEdit: A Unified Image Editing Dataset and Benchmark","ref_index":62,"is_internal_anchor":false},{"citing_arxiv_id":"2506.13585","citing_title":"MiniMax-M1: Scaling Test-Time Compute Efficiently with Lightning Attention","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06996","citing_title":"Self-Preference Bias in Rubric-Based Evaluation of Large Language Models","ref_index":14,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YOSF547IIPN4WTGWEVOA2WM777","json":"https://pith.science/pith/YOSF547IIPN4WTGWEVOA2WM777.json","graph_json":"https://pith.science/api/pith-number/YOSF547IIPN4WTGWEVOA2WM777/graph.json","events_json":"https://pith.science/api/pith-number/YOSF547IIPN4WTGWEVOA2WM777/events.json","paper":"https://pith.science/paper/YOSF547I"},"agent_actions":{"view_html":"https://pith.science/pith/YOSF547IIPN4WTGWEVOA2WM777","download_json":"https://pith.science/pith/YOSF547IIPN4WTGWEVOA2WM777.json","view_paper":"https://pith.science/paper/YOSF547I","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.17399&json=true","fetch_graph":"https://pith.science/api/pith-number/YOSF547IIPN4WTGWEVOA2WM777/graph.json","fetch_events":"https://pith.science/api/pith-number/YOSF547IIPN4WTGWEVOA2WM777/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YOSF547IIPN4WTGWEVOA2WM777/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YOSF547IIPN4WTGWEVOA2WM777/action/storage_attestation","attest_author":"https://pith.science/pith/YOSF547IIPN4WTGWEVOA2WM777/action/author_attestation","sign_citation":"https://pith.science/pith/YOSF547IIPN4WTGWEVOA2WM777/action/citation_signature","submit_replication":"https://pith.science/pith/YOSF547IIPN4WTGWEVOA2WM777/action/replication_record"}},"created_at":"2026-07-05T10:25:16.631172+00:00","updated_at":"2026-07-05T10:25:16.631172+00:00"}