{
  "schema": "lri.security-paper-metadata-and-upgrade-blockers.v1",
  "metadata_count": 12,
  "blocker_count": 3,
  "paper_metadata": [
    {
      "card_id": "molot_system_card",
      "paper_title": "MOLOT System Card: Malicious Operational Logic Observation Transformer",
      "arxiv_id": "2606.07792",
      "primary_url": "https://arxiv.org/abs/2606.07792",
      "artifact_urls": [],
      "directness": "direct_system_card",
      "verified_dimensions": [
        "SAST setting",
        "static call graph",
        "behavior sequence",
        "SHAP source-location mapping",
        "Python/JavaScript package evaluation",
        "product constraints"
      ],
      "missing_dimensions": [
        "independent Open Malicious-Code Bench URL",
        "full benchmark table extraction",
        "runtime/memory/FPR numeric extraction"
      ],
      "promotion_level": "L2_direct_public_evidence"
    },
    {
      "card_id": "cerebro_behavior_sequence",
      "paper_title": "Malicious Package Detection in NPM and PyPI using a Single Model of Malicious Behavior Sequence",
      "arxiv_id": "2309.02637",
      "primary_url": "https://arxiv.org/abs/2309.02637",
      "artifact_urls": [],
      "directness": "method_background",
      "verified_dimensions": [
        "NPM/PyPI malicious package detection",
        "high-level malicious behavior abstraction",
        "behavior sequence",
        "BERT fine-tuning"
      ],
      "missing_dimensions": [
        "mapping table to MOLOT activity taxonomy",
        "replication artifact URL"
      ],
      "promotion_level": "L1_public_method_background"
    },
    {
      "card_id": "python_malware_benchmark",
      "paper_title": "A Benchmark Comparison of Python Malware Detection Approaches",
      "arxiv_id": "2209.13288",
      "primary_url": "https://arxiv.org/abs/2209.13288",
      "artifact_urls": [],
      "directness": "baseline_benchmark",
      "verified_dimensions": [
        "PyPI malware benchmark",
        "PyPI checks / Bandit4Mal / OSSGadget comparison",
        "false-positive constraint"
      ],
      "missing_dimensions": [
        "exact benchmark dataset URL",
        "table extraction into structured rows"
      ],
      "promotion_level": "L2_direct_baseline_evidence"
    },
    {
      "card_id": "npm_malicious_detection_benchmark",
      "paper_title": "Understanding NPM Malicious Package Detection: A Benchmark-Driven Empirical Analysis",
      "arxiv_id": "2603.27549",
      "primary_url": "https://arxiv.org/abs/2603.27549",
      "artifact_urls": [],
      "directness": "ecosystem_benchmark",
      "verified_dimensions": [
        "NPM malicious/benign package dataset",
        "11 behavior categories",
        "8 evasion techniques",
        "8 tools / 13 variants",
        "behavioral-chain finding"
      ],
      "missing_dimensions": [
        "public dataset/framework repository URL",
        "extract tool-combination table"
      ],
      "promotion_level": "L2_direct_benchmark_evidence"
    },
    {
      "card_id": "oss_malicious_package_analysis",
      "paper_title": "OSS Malicious Package Analysis in the Wild",
      "arxiv_id": "2404.04991",
      "primary_url": "https://arxiv.org/abs/2404.04991",
      "artifact_urls": [],
      "directness": "dataset_lifecycle_analysis",
      "verified_dimensions": [
        "23,425 malicious packages",
        "knowledge graph",
        "source overlap issue",
        "lifecycle changing-release-detection-removal"
      ],
      "missing_dimensions": [
        "dataset download URL",
        "schema fields extraction"
      ],
      "promotion_level": "L2_direct_dataset_evidence"
    },
    {
      "card_id": "backstabbers_knife_collection",
      "paper_title": "Backstabber's Knife Collection",
      "arxiv_id": "2005.09535",
      "primary_url": "https://arxiv.org/abs/2005.09535",
      "artifact_urls": [],
      "directness": "supply_chain_taxonomy",
      "verified_dimensions": [
        "174 malicious packages",
        "npm/PyPI/RubyGems",
        "attack trees",
        "execution timing"
      ],
      "missing_dimensions": [
        "machine-readable dataset URL",
        "execution phase mapping"
      ],
      "promotion_level": "L1_public_supply_chain_taxonomy"
    },
    {
      "card_id": "unveiling_malicious_logic",
      "paper_title": "Unveiling Malicious Logic",
      "arxiv_id": "2512.12559",
      "primary_url": "https://arxiv.org/abs/2512.12559",
      "artifact_urls": [],
      "directness": "statement_level_taxonomy",
      "verified_dimensions": [
        "statement-level dataset",
        "370 malicious Python packages",
        "47 indicators / 7 types",
        "sequential pattern mining"
      ],
      "missing_dimensions": [
        "indicator taxonomy extraction",
        "dataset artifact URL"
      ],
      "promotion_level": "L2_direct_statement_level_evidence"
    },
    {
      "card_id": "llm_pypi_malicious_detection",
      "paper_title": "Detecting Malicious Source Code in PyPI Packages with LLMs",
      "arxiv_id": "2504.13769",
      "primary_url": "https://arxiv.org/abs/2504.13769",
      "artifact_urls": [],
      "directness": "llm_labeling_reference",
      "verified_dimensions": [
        "LLM/RAG/few-shot PyPI detection evaluation",
        "YARA/GitHub Security Advisories/snippet integration",
        "RAG modest, few-shot stronger"
      ],
      "missing_dimensions": [
        "prompt artifact extraction",
        "dataset/code URL"
      ],
      "promotion_level": "L1_public_llm_labeling_evidence"
    },
    {
      "card_id": "pdg_holmes_graph_siamese",
      "paper_title": "Modeling Functional Similarity in Source Code with Graph-Based Siamese Networks",
      "arxiv_id": "2011.11228",
      "primary_url": "https://arxiv.org/abs/2011.11228",
      "artifact_urls": [],
      "directness": "graph_clone_background",
      "verified_dimensions": [
        "program dependency graphs",
        "semantic clone detection",
        "graph-based Siamese HOLMES",
        "clone benchmark evaluation"
      ],
      "missing_dimensions": [
        "HOLMES artifact URL",
        "relation to FCD table"
      ],
      "promotion_level": "L1_public_graph_clone_background"
    },
    {
      "card_id": "source_code_is_graph",
      "paper_title": "Source Code is a Graph, Not a Sequence",
      "arxiv_id": "2312.16488",
      "primary_url": "https://arxiv.org/abs/2312.16488",
      "artifact_urls": [],
      "directness": "graph_vs_sequence_clone_background",
      "verified_dimensions": [
        "graph-based code representation",
        "CodeGraph vs CodeBERT",
        "cross-lingual clone detection"
      ],
      "missing_dimensions": [
        "benchmark artifact URLs",
        "relation to binary clone detection"
      ],
      "promotion_level": "L1_public_graph_clone_background"
    },
    {
      "card_id": "shap_unified_approach",
      "paper_title": "A Unified Approach to Interpreting Model Predictions",
      "arxiv_id": "1705.07874",
      "primary_url": "https://arxiv.org/abs/1705.07874",
      "artifact_urls": [],
      "directness": "explainability_foundation",
      "verified_dimensions": [
        "SHAP framework",
        "feature attribution",
        "additive feature importance"
      ],
      "missing_dimensions": [
        "code-localization mapping, supplied by MOLOT not SHAP itself"
      ],
      "promotion_level": "L1_public_explainability_foundation"
    },
    {
      "card_id": "fcd_pdg_performance_unresolved",
      "paper_title": "FCD / PDG C/C++ Performance Table from User JSON",
      "arxiv_id": null,
      "primary_url": "user_provided_json",
      "artifact_urls": [],
      "directness": "unresolved_user_claim",
      "verified_dimensions": [
        "none_public"
      ],
      "missing_dimensions": [
        "original paper/report URL",
        "definition of FCD",
        "dataset/task setup",
        "relation to PDG/HOLMES/CodeGraph/MOLOT"
      ],
      "promotion_level": "L0_user_input_only"
    }
  ],
  "upgrade_blockers": [
    {
      "id": "block_fcd_pdg_table_upgrade",
      "claim": "FCD/PDG C/C++ table precision/recall/RMSE/FCD speed is verified.",
      "current_status": "blocked",
      "reason": "No direct public original source URL was found for the exact table/project-version rows.",
      "required_evidence": [
        "original paper/report URL",
        "table screenshot or text in source",
        "definition of FCD",
        "task/dataset description",
        "author/system linkage"
      ],
      "currently_available": [
        "user_provided_json",
        "PDG/HOLMES graph-clone background",
        "CodeGraph clone-detection background"
      ],
      "allowed_wording": "待核验用户线索；可能属于源码/二进制相似片段检测方向。"
    },
    {
      "id": "block_open_malicious_code_bench_artifact_upgrade",
      "claim": "Open Malicious-Code Bench independent artifact URL is verified.",
      "current_status": "blocked",
      "reason": "Search found the MOLOT paper statement, but no stable independent project/GitHub URL in this iteration.",
      "required_evidence": [
        "independent repository/project URL",
        "benchmark schema or README",
        "download/access instructions"
      ],
      "currently_available": [
        "MOLOT arXiv statement that the benchmark is released"
      ],
      "allowed_wording": "论文声明释放 Open Malicious-Code Bench；独立 artifact URL 待补。"
    },
    {
      "id": "block_agent_gate_performance_upgrade",
      "claim": "LRI Agent Gate has verified precision/latency/FPR.",
      "current_status": "blocked",
      "reason": "No LRI replayable tests, audit logs, latency or false-positive measurements have been produced yet.",
      "required_evidence": [
        "MVP test execution logs",
        "scanner version",
        "sample set",
        "false positive/false negative review",
        "latency/memory measurement"
      ],
      "currently_available": [
        "security_gate_contract",
        "mvp_test_cases",
        "reviewer_evidence_schema"
      ],
      "allowed_wording": "工程设计候选，等待 MVP 测试验证。"
    }
  ]
}