{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:3TILJDF3UTLXXX447Z5VZGW2F7","short_pith_number":"pith:3TILJDF3","schema_version":"1.0","canonical_sha256":"dcd0b48cbba4d77bdf9cfe7b5c9ada2ff5a5619daf6adb449f353b1ec4ed902b","source":{"kind":"arxiv","id":"2410.15553","version":2},"attestation_state":"computed","paper":{"title":"Multi-IF: Benchmarking LLMs on Multi-Turn and Multilingual Instructions Following","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Aditya Tayade, Chaoqi Wang, Chenguang Zhu, Chen Zhu, Chloe Bi, Di Jin, Eryk Helenowski, Han Fang, Hao Ma, Hejia Zhang, Hongjiang Lv, Karishma Mandyam, Karthik Abinav Sankararaman, Melanie Kambadur, Ning Li, Shruti Bhosale, Sinong Wang, Tengyu Xu, Yun He","submitted_at":"2024-10-21T00:59:47Z","abstract_excerpt":"Large Language Models (LLMs) have demonstrated impressive capabilities in various tasks, including instruction following, which is crucial for aligning model outputs with user expectations. However, evaluating LLMs' ability to follow instructions remains challenging due to the complexity and subjectivity of human language. Current benchmarks primarily focus on single-turn, monolingual instructions, which do not adequately reflect the complexities of real-world applications that require handling multi-turn and multilingual interactions. To address this gap, we introduce Multi-IF, a new benchmar"},"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":"2410.15553","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-10-21T00:59:47Z","cross_cats_sorted":[],"title_canon_sha256":"cff093408af0fb3e084a34420e529e7675f97130dc708a75265f3fbcbe41cda8","abstract_canon_sha256":"1cebc88d6bc28ec8be5d5d44e06979224c92efa65d73ba3bdd3b6cb9de3f48c8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:34:38.927837Z","signature_b64":"ktlAmT9V4FO4eOffGCBWWFU1tQi3/74/CzZXbFzTBMWeD+BdlWXCL8nqFp3bVLjFuXTwAyxBBN4HiB7fHBVSDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"dcd0b48cbba4d77bdf9cfe7b5c9ada2ff5a5619daf6adb449f353b1ec4ed902b","last_reissued_at":"2026-07-05T09:34:38.927344Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:34:38.927344Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Multi-IF: Benchmarking LLMs on Multi-Turn and Multilingual Instructions Following","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Aditya Tayade, Chaoqi Wang, Chenguang Zhu, Chen Zhu, Chloe Bi, Di Jin, Eryk Helenowski, Han Fang, Hao Ma, Hejia Zhang, Hongjiang Lv, Karishma Mandyam, Karthik Abinav Sankararaman, Melanie Kambadur, Ning Li, Shruti Bhosale, Sinong Wang, Tengyu Xu, Yun He","submitted_at":"2024-10-21T00:59:47Z","abstract_excerpt":"Large Language Models (LLMs) have demonstrated impressive capabilities in various tasks, including instruction following, which is crucial for aligning model outputs with user expectations. However, evaluating LLMs' ability to follow instructions remains challenging due to the complexity and subjectivity of human language. Current benchmarks primarily focus on single-turn, monolingual instructions, which do not adequately reflect the complexities of real-world applications that require handling multi-turn and multilingual interactions. To address this gap, we introduce Multi-IF, a new benchmar"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.15553","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/2410.15553/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":"2410.15553","created_at":"2026-07-05T09:34:38.927404+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.15553v2","created_at":"2026-07-05T09:34:38.927404+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.15553","created_at":"2026-07-05T09:34:38.927404+00:00"},{"alias_kind":"pith_short_12","alias_value":"3TILJDF3UTLX","created_at":"2026-07-05T09:34:38.927404+00:00"},{"alias_kind":"pith_short_16","alias_value":"3TILJDF3UTLXXX44","created_at":"2026-07-05T09:34:38.927404+00:00"},{"alias_kind":"pith_short_8","alias_value":"3TILJDF3","created_at":"2026-07-05T09:34:38.927404+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":17,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.15079","citing_title":"Ling and Ring 2.6 Technical Report: Efficient and Instant Agentic Intelligence at Trillion-Parameter Scale","ref_index":149,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22064","citing_title":"Hy-MT2: A Family of Fast, Efficient and Powerful Multilingual Translation Models in the Wild","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2605.28218","citing_title":"IFMTBench: A Comprehensive Benchmark for Multilingual Translation Instruction Following","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22064","citing_title":"Hy-MT2: A Family of Fast, Efficient and Powerful Multilingual Translation Models in the Wild","ref_index":82,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07731","citing_title":"Benchmarking EngGPT2-16B-A3B against Comparable Italian and International Open-source LLMs","ref_index":70,"is_internal_anchor":false},{"citing_arxiv_id":"2601.20251","citing_title":"Efficient Evaluation of LLM Performance with Statistical Guarantees","ref_index":10,"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":130,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12922","citing_title":"When Attention Closes: How LLMs Lose the Thread in Multi-Turn Interaction","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2505.20275","citing_title":"ImgEdit: A Unified Image Editing Dataset and Benchmark","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2502.01456","citing_title":"Process Reinforcement through Implicit Rewards","ref_index":125,"is_internal_anchor":false},{"citing_arxiv_id":"2604.21827","citing_title":"Alignment has a Fantasia Problem","ref_index":68,"is_internal_anchor":false},{"citing_arxiv_id":"2604.09285","citing_title":"SAGE: A Service Agent Graph-guided Evaluation Benchmark","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07465","citing_title":"SEIF: Self-Evolving Reinforcement Learning for Instruction Following","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07731","citing_title":"Benchmarking EngGPT2-16B-A3B against Comparable Italian and International Open-source LLMs","ref_index":70,"is_internal_anchor":false},{"citing_arxiv_id":"2509.17765","citing_title":"Qwen3-Omni Technical Report","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2509.17765","citing_title":"Qwen3-Omni Technical Report","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2505.09388","citing_title":"Qwen3 Technical Report","ref_index":17,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3TILJDF3UTLXXX447Z5VZGW2F7","json":"https://pith.science/pith/3TILJDF3UTLXXX447Z5VZGW2F7.json","graph_json":"https://pith.science/api/pith-number/3TILJDF3UTLXXX447Z5VZGW2F7/graph.json","events_json":"https://pith.science/api/pith-number/3TILJDF3UTLXXX447Z5VZGW2F7/events.json","paper":"https://pith.science/paper/3TILJDF3"},"agent_actions":{"view_html":"https://pith.science/pith/3TILJDF3UTLXXX447Z5VZGW2F7","download_json":"https://pith.science/pith/3TILJDF3UTLXXX447Z5VZGW2F7.json","view_paper":"https://pith.science/paper/3TILJDF3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.15553&json=true","fetch_graph":"https://pith.science/api/pith-number/3TILJDF3UTLXXX447Z5VZGW2F7/graph.json","fetch_events":"https://pith.science/api/pith-number/3TILJDF3UTLXXX447Z5VZGW2F7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3TILJDF3UTLXXX447Z5VZGW2F7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3TILJDF3UTLXXX447Z5VZGW2F7/action/storage_attestation","attest_author":"https://pith.science/pith/3TILJDF3UTLXXX447Z5VZGW2F7/action/author_attestation","sign_citation":"https://pith.science/pith/3TILJDF3UTLXXX447Z5VZGW2F7/action/citation_signature","submit_replication":"https://pith.science/pith/3TILJDF3UTLXXX447Z5VZGW2F7/action/replication_record"}},"created_at":"2026-07-05T09:34:38.927404+00:00","updated_at":"2026-07-05T09:34:38.927404+00:00"}