{
  "aliases": [],
  "authors": [],
  "details": {
    "availability": "Public federated-training/evaluation code and ungated annotation partitions; original images are separate dependencies.",
    "benchmark_protocol": {
      "actions": [
        "model predictions on mobile-agent examples"
      ],
      "counts": [
        {
          "basis": "reported",
          "evidence_url": "https://huggingface.co/datasets/wwh0411/FedMABench/raw/main/README.md",
          "population": "released FedMABench partition families",
          "qualifier": "exact",
          "unit": "dataset families",
          "value": 6
        },
        {
          "basis": "reported",
          "evidence_url": "https://raw.githubusercontent.com/wwh0411/FedMABench/HEAD/README.md",
          "population": "publisher benchmark overview",
          "qualifier": "exact",
          "unit": "heterogeneity types",
          "value": 3
        },
        {
          "basis": "reported",
          "evidence_url": "https://raw.githubusercontent.com/wwh0411/FedMABench/HEAD/README.md",
          "population": "publisher benchmark overview",
          "qualifier": "exact",
          "unit": "federated learning algorithms",
          "value": 8
        }
      ],
      "evaluation": "The provided evaluation workflow performs model inference over selected partitions and invokes test_swift_fed.py for accuracy scoring.",
      "evidence_urls": [
        "https://raw.githubusercontent.com/wwh0411/FedMABench/HEAD/README.md",
        "https://huggingface.co/datasets/wwh0411/FedMABench/raw/main/README.md"
      ],
      "focus": [
        "mobile-agent federated training",
        "data heterogeneity"
      ],
      "notes": "The dataset-family count does not count tasks, apps, screenshots or client instances. Code and data licenses are unknown; external image sources are required.",
      "observations": [
        "mobile-agent examples with externally sourced screenshots"
      ],
      "setting": "offline"
    },
    "repository_revision": "417d8b0006830f423c7d07c213fc57277d75d966"
  },
  "environments": [
    "android"
  ],
  "evidence": [
    {
      "method": "editorial_primary_review",
      "retrieved_at": "2026-10-17T11:18:17Z",
      "role": "primary",
      "sha256": "b49e2b419918078ef9c68f3c9becc6c13d8fe530c165a36bdd61e370746eeea8",
      "supports": [
        "name",
        "kind",
        "summary",
        "environments",
        "topics",
        "scope",
        "notes",
        "relations",
        "details.benchmark_protocol"
      ],
      "url": "https://raw.githubusercontent.com/wwh0411/FedMABench/HEAD/README.md"
    },
    {
      "method": "editorial_primary_review",
      "retrieved_at": "2026-10-17T11:18:17Z",
      "role": "primary",
      "sha256": "3e1b2b643a263af129b3a4092a84efe47d3643f00509f7b9517da21a4aac0371",
      "supports": [
        "identity"
      ],
      "url": "https://api.github.com/repos/wwh0411/FedMABench"
    },
    {
      "method": "editorial_primary_review",
      "retrieved_at": "2026-10-17T11:22:54Z",
      "role": "primary",
      "sha256": "4fb977a469cee3d5083da0b5bf2984fedc6b867146a90b20963985e08f9e9716",
      "supports": [
        "release_status",
        "details.repository_revision",
        "licenses.code"
      ],
      "url": "https://api.github.com/repos/wwh0411/FedMABench/git/trees/master?recursive=1"
    },
    {
      "method": "editorial_primary_review",
      "retrieved_at": "2026-10-17T11:23:06Z",
      "role": "primary",
      "sha256": "df5379f4a9b350dad0c3d374d2f9a1d6eacd974b6d54750e2a14ecec1faec6e5",
      "supports": [
        "identity",
        "release_status"
      ],
      "url": "https://huggingface.co/api/datasets/wwh0411/FedMABench"
    },
    {
      "method": "editorial_primary_review",
      "retrieved_at": "2026-10-17T11:23:06Z",
      "role": "primary",
      "sha256": "2874269995d8c9f5dabe266433ecf724bf54c27af599d899f3f0090b8ee37cdc",
      "supports": [
        "summary",
        "details.benchmark_protocol",
        "licenses.data"
      ],
      "url": "https://huggingface.co/datasets/wwh0411/FedMABench/raw/main/README.md"
    }
  ],
  "id": "github:wwh0411/fedmabench",
  "identifiers": {},
  "kind": "benchmark",
  "last_verified_at": "2026-10-17T11:18:17Z",
  "licenses": {
    "code": null,
    "data": null,
    "document": null,
    "weights": null
  },
  "links": [
    {
      "rel": "primary",
      "url": "https://github.com/wwh0411/FedMABench"
    }
  ],
  "name": "FedMABench",
  "notes": "The public Hub lists six dataset families of JSON/JSONL partition annotations. The repository explicitly says unpacked screenshots are not provided and must be reconstructed from original AndroidControl and AITW sources. This is offline training/evaluation infrastructure rather than a live Android task environment. No repository license file or dataset-license declaration was found in the inspected sources; neither is inferred from the underlying ms-swift project.",
  "published_at": null,
  "relations": [
    {
      "evidence_url": "https://raw.githubusercontent.com/wwh0411/FedMABench/HEAD/README.md",
      "target_id": "arxiv:2503.05143",
      "type": "documented_by"
    }
  ],
  "review_status": "curated",
  "scope": "core",
  "summary": "A federated mobile-agent training and evaluation suite that varies user-data heterogeneity and application distribution, with released partition annotations and inference/scoring scripts.",
  "topics": [
    "evaluation",
    "imitation-learning",
    "mobile-navigation"
  ],
  "updated_at": null,
  "url": "https://github.com/wwh0411/FedMABench"
}