{"record": {"aliases": ["COLLIE"], "artifacts": [{"kind": "website", "url": "https://collie-benchmark.github.io/"}], "audits": [], "categories": ["reasoning"], "description": {"en": "Systematic construction of constrained text generation tasks; evaluates rich, compositional constraints across diverse generation levels."}, "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:collie", "modality": null, "name": "COLLIE", "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://collie-benchmark.github.io/"}, "publisher": null, "released": "2023-07-17", "released_reference": {"basis": "paper_first_version", "note": "First version of the introducing paper: COLLIE: Systematic Construction of Constrained Text Generation Tasks. The source-record description identifies this benchmark or its original partition; no dates or scores are joined by name at runtime.", "source_key": "model-reports:collie", "source_url": "https://arxiv.org/abs/2307.08689"}, "schema_version": 1, "sizes": [], "slug": "collie", "source": "model_reports", "source_benchmark_id": "collie"}, "schema_version": 1, "scores_by_source": {"model_reports": {"rows": [{"benchmark_id": "collie", "comparable_group": "[\"model-reports:collie\",\"collie\",\"as printed in Table 3 (Large Models)\"]", "date_precision": "document_publication", "document_id": "model_reports:bytedance_seed2_0_model_card", "instrument": "collie", "key": "model-reports:collie", "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\u0000collie\u0000bytedance_seed2_0_model_card\u0000collie\u0000as printed in Table 3 (Large Models)\u0000Seed2.0 Pro", "observation_id": "curated\u0000collie\u0000bytedance_seed2_0_model_card\u0000collie\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": "93.9", "read_from": "pdf_text", "reported_at": "2026-02-14", "reported_by": "self_reported", "reported_date": "2026-02-14", "series_id": "model_reports:collie: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": 93.9, "value_kind": "number"}, {"benchmark_id": "collie", "comparable_group": "[\"model-reports:collie\",\"collie\",\"as printed in Table 4 (Efficient Models)\"]", "date_precision": "document_publication", "document_id": "model_reports:bytedance_seed2_0_model_card", "instrument": "collie", "key": "model-reports:collie", "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\u0000collie\u0000bytedance_seed2_0_model_card\u0000collie\u0000as printed in Table 4 (Efficient Models)\u0000Seed2.0 Lite", "observation_id": "curated\u0000collie\u0000bytedance_seed2_0_model_card\u0000collie\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": "94.0", "read_from": "pdf_text", "reported_at": "2026-02-14", "reported_by": "self_reported", "reported_date": "2026-02-14", "series_id": "model_reports:collie: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": 94.0, "value_kind": "number"}, {"benchmark_id": "collie", "comparable_group": "[\"model-reports:collie\",\"collie\",\"as printed in Table 4 (Efficient Models)\"]", "date_precision": "document_publication", "document_id": "model_reports:bytedance_seed2_0_model_card", "instrument": "collie", "key": "model-reports:collie", "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\u0000collie\u0000bytedance_seed2_0_model_card\u0000collie\u0000as printed in Table 4 (Efficient Models)\u0000Seed2.0 Mini", "observation_id": "curated\u0000collie\u0000bytedance_seed2_0_model_card\u0000collie\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": "91.2", "read_from": "pdf_text", "reported_at": "2026-02-14", "reported_by": "self_reported", "reported_date": "2026-02-14", "series_id": "model_reports:collie: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": 91.2, "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\u0000collie\u0000bytedance_seed2_0_model_card\u0000collie\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\u0000collie\u0000bytedance_seed2_0_model_card\u0000collie\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:collie", "max_score_contradicted": false, "metric": "accuracy", "observation_count": 3, "observed_max": 94.0, "score_summary": {"display_max": 94.0, "display_multiplier": 1, "model_count": 3, "model_count_basis": "source_model_id", "numeric_count": 3, "raw_max": 94.0, "raw_min": 91.2, "source_reference": {"obs_id": "curated\u0000collie\u0000bytedance_seed2_0_model_card\u0000collie\u0000as printed in Table 4 (Efficient Models)\u0000Seed2.0 Lite", "observation_id": "curated\u0000collie\u0000bytedance_seed2_0_model_card\u0000collie\u0000as printed in Table 4 (Efficient Models)\u0000Seed2.0 Lite", "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:collie:default", "unit": "percent"}}}, "siblings": []}
