{"record": {"aliases": ["Inverse IFEval"], "artifacts": [{"kind": "paper", "url": "https://arxiv.org/abs/2509.04292"}], "audits": [], "authors": ["Qinyan Zhang", "Xinping Lei", "Ruijie Miao", "Yu Fu", "Haojie Fan", "Le Chang", "Jiafan Hou", "Dingling Zhang", "Zhongfei Hou", "Ziqiang Yang", "Changxin Pu", "Fei Hu", "Jingkai Liu", "Mengyun Liu", "Yang Liu", "Xiang Gao", "Jiaheng Liu", "Tong Yang", "Zaiyuan Wang", "Ge Zhang", "Wenhao Huang"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "model-reports:inverse_ifeval", "source_url": "https://arxiv.org/abs/2509.04292"}, "categories": ["instruction_following"], "description": {"en": "Tests ability to override training biases and follow counterintuitive instructions; complementary to IFEval."}, "documents": [{"document_type": "model_card", "id": "model_reports:bytedance_seed2_0_model_card", "model_name": "Seed2.0 (Pro / Lite / Mini)", "organization": "ByteDance", "published": "2026-02-14", "retrieved_at": "2026-09-11", "source": "model_reports", "source_id": "bytedance_seed2_0_model_card", "source_url": "https://lf3-static.bytednsdoc.com/obj/eden-cn/lapzild-tss/ljhwZthlaukjlkulzlp/seed2/0214/Seed2.0%20Model%20Card.pdf", "title": "Seed2.0 (Pro / Lite / Mini)"}], "evidence_summary": {"document_count": 1, "model_count": 3, "model_count_basis": "source_model_id"}, "key": "model-reports:inverse_ifeval", "modality": null, "name": "Inverse IFEval", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"registry": "data/model_cards.yml", "score_archive": "data/benchmark_scores.yml", "source_url": "https://arxiv.org/abs/2509.04292"}, "publisher": null, "released": "2025-09-04", "released_reference": {"basis": "benchmark_release", "source_key": "model-reports:inverse_ifeval", "source_url": "https://arxiv.org/abs/2509.04292"}, "schema_version": 1, "sizes": [], "slug": "inverse_ifeval", "source": "model_reports", "source_benchmark_id": "inverse_ifeval"}, "schema_version": 1, "scores_by_source": {"model_reports": {"rows": [{"benchmark_id": "inverse_ifeval", "comparable_group": "[\"model-reports:inverse_ifeval\",\"inverse_ifeval\",\"as printed in Table 3 (Large Models)\"]", "date_precision": "document_publication", "document_id": "model_reports:bytedance_seed2_0_model_card", "instrument": "inverse_ifeval", "key": "model-reports:inverse_ifeval", "measured_by": "ByteDance", "measurement_kind": "reported_self_score", "model": "Seed2.0 Pro", "model_id": "[\"ByteDance\",\"Seed2.0 Pro\"]", "model_name": "Seed2.0 Pro", "obs_id": "curated\u0000inverse_ifeval\u0000bytedance_seed2_0_model_card\u0000inverse_ifeval\u0000as printed in Table 3 (Large Models)\u0000Seed2.0 Pro", "observation_id": "curated\u0000inverse_ifeval\u0000bytedance_seed2_0_model_card\u0000inverse_ifeval\u0000as printed in Table 3 (Large Models)\u0000Seed2.0 Pro", "organization": "ByteDance", "protocol": "as printed in Table 3 (Large Models)", "rank_in_source_response": null, "raw_value": "78.9", "read_from": "pdf_text", "reported_at": "2026-02-14", "reported_by": "self_reported", "reported_date": "2026-02-14", "series_id": "model_reports:inverse_ifeval:default", "source": "model_reports", "source_document_type": "model_card", "source_id": "bytedance_seed2_0_model_card", "source_title": "Seed2.0 (Pro / Lite / Mini)", "source_url": "https://lf3-static.bytednsdoc.com/obj/eden-cn/lapzild-tss/ljhwZthlaukjlkulzlp/seed2/0214/Seed2.0%20Model%20Card.pdf", "value": 78.9, "value_kind": "number"}, {"benchmark_id": "inverse_ifeval", "comparable_group": "[\"model-reports:inverse_ifeval\",\"inverse_ifeval\",\"as printed in Table 4 (Efficient Models)\"]", "date_precision": "document_publication", "document_id": "model_reports:bytedance_seed2_0_model_card", "instrument": "inverse_ifeval", "key": "model-reports:inverse_ifeval", "measured_by": "ByteDance", "measurement_kind": "reported_self_score", "model": "Seed2.0 Lite", "model_id": "[\"ByteDance\",\"Seed2.0 Lite\"]", "model_name": "Seed2.0 Lite", "obs_id": "curated\u0000inverse_ifeval\u0000bytedance_seed2_0_model_card\u0000inverse_ifeval\u0000as printed in Table 4 (Efficient Models)\u0000Seed2.0 Lite", "observation_id": "curated\u0000inverse_ifeval\u0000bytedance_seed2_0_model_card\u0000inverse_ifeval\u0000as printed in Table 4 (Efficient Models)\u0000Seed2.0 Lite", "organization": "ByteDance", "protocol": "as printed in Table 4 (Efficient Models)", "rank_in_source_response": null, "raw_value": "77.1", "read_from": "pdf_text", "reported_at": "2026-02-14", "reported_by": "self_reported", "reported_date": "2026-02-14", "series_id": "model_reports:inverse_ifeval:default", "source": "model_reports", "source_document_type": "model_card", "source_id": "bytedance_seed2_0_model_card", "source_title": "Seed2.0 (Pro / Lite / Mini)", "source_url": "https://lf3-static.bytednsdoc.com/obj/eden-cn/lapzild-tss/ljhwZthlaukjlkulzlp/seed2/0214/Seed2.0%20Model%20Card.pdf", "value": 77.1, "value_kind": "number"}, {"benchmark_id": "inverse_ifeval", "comparable_group": "[\"model-reports:inverse_ifeval\",\"inverse_ifeval\",\"as printed in Table 4 (Efficient Models)\"]", "date_precision": "document_publication", "document_id": "model_reports:bytedance_seed2_0_model_card", "instrument": "inverse_ifeval", "key": "model-reports:inverse_ifeval", "measured_by": "ByteDance", "measurement_kind": "reported_self_score", "model": "Seed2.0 Mini", "model_id": "[\"ByteDance\",\"Seed2.0 Mini\"]", "model_name": "Seed2.0 Mini", "obs_id": "curated\u0000inverse_ifeval\u0000bytedance_seed2_0_model_card\u0000inverse_ifeval\u0000as printed in Table 4 (Efficient Models)\u0000Seed2.0 Mini", "observation_id": "curated\u0000inverse_ifeval\u0000bytedance_seed2_0_model_card\u0000inverse_ifeval\u0000as printed in Table 4 (Efficient Models)\u0000Seed2.0 Mini", "organization": "ByteDance", "protocol": "as printed in Table 4 (Efficient Models)", "rank_in_source_response": null, "raw_value": "69.3", "read_from": "pdf_text", "reported_at": "2026-02-14", "reported_by": "self_reported", "reported_date": "2026-02-14", "series_id": "model_reports:inverse_ifeval:default", "source": "model_reports", "source_document_type": "model_card", "source_id": "bytedance_seed2_0_model_card", "source_title": "Seed2.0 (Pro / Lite / Mini)", "source_url": "https://lf3-static.bytednsdoc.com/obj/eden-cn/lapzild-tss/ljhwZthlaukjlkulzlp/seed2/0214/Seed2.0%20Model%20Card.pdf", "value": 69.3, "value_kind": "number"}], "series": {"declared_max": null, "direction": "higher_is_better", "direction_basis": "reported_metric", "display_scale": null, "first_score_record": {"date_precision": "document_publication", "obs_id": "curated\u0000inverse_ifeval\u0000bytedance_seed2_0_model_card\u0000inverse_ifeval\u0000as printed in Table 3 (Large Models)\u0000Seed2.0 Pro", "reported_at": "2026-02-14", "source_url": "https://lf3-static.bytednsdoc.com/obj/eden-cn/lapzild-tss/ljhwZthlaukjlkulzlp/seed2/0214/Seed2.0%20Model%20Card.pdf"}, "first_score_report": {"date_precision": "document_publication", "obs_id": "curated\u0000inverse_ifeval\u0000bytedance_seed2_0_model_card\u0000inverse_ifeval\u0000as printed in Table 3 (Large Models)\u0000Seed2.0 Pro", "reported_at": "2026-02-14", "source_url": "https://lf3-static.bytednsdoc.com/obj/eden-cn/lapzild-tss/ljhwZthlaukjlkulzlp/seed2/0214/Seed2.0%20Model%20Card.pdf"}, "key": "model-reports:inverse_ifeval", "max_score_contradicted": false, "metric": "accuracy", "observation_count": 3, "observed_max": 78.9, "score_summary": {"display_max": 78.9, "display_multiplier": 1, "model_count": 3, "model_count_basis": "source_model_id", "numeric_count": 3, "raw_max": 78.9, "raw_min": 69.3, "source_reference": {"obs_id": "curated\u0000inverse_ifeval\u0000bytedance_seed2_0_model_card\u0000inverse_ifeval\u0000as printed in Table 3 (Large Models)\u0000Seed2.0 Pro", "observation_id": "curated\u0000inverse_ifeval\u0000bytedance_seed2_0_model_card\u0000inverse_ifeval\u0000as printed in Table 3 (Large Models)\u0000Seed2.0 Pro", "reported_at": "2026-02-14", "reported_date": "2026-02-14", "source_id": "bytedance_seed2_0_model_card", "source_url": "https://lf3-static.bytednsdoc.com/obj/eden-cn/lapzild-tss/ljhwZthlaukjlkulzlp/seed2/0214/Seed2.0%20Model%20Card.pdf"}, "unit": "percent"}, "series_id": "model_reports:inverse_ifeval:default", "unit": "percent"}}}, "siblings": []}
