logbook.json

{
  "schema_version": 2,
  "title": "Repro - NonZero: Interaction-Guided Exploration for Multi-Agent Monte Carlo Tree Search",
  "emoji": "🎯",
  "space_id": "byte-vortex/nonzero-repro",
  "paper": {
    "arxiv_id": "2605.00751"
  },
  "tags": [
    "icml2026-repro",
    "paper-Jh6gq9QsFa"
  ],
  "updated_at": "2026-07-27T04:27:10+00:00",
  "root": {
    "slug": "index",
    "title": "Repro - NonZero: Interaction-Guided Exploration for Multi-Agent Monte Carlo Tree Search",
    "file": "pages/index.md",
    "children": [
      {
        "slug": "claim-1-low-dimensional-nonlinear-surrogate",
        "title": "Claim 1: Low-dimensional nonlinear surrogate",
        "file": "pages/claim-1-low-dimensional-nonlinear-surrogate/page.md",
        "children": []
      },
      {
        "slug": "claim-2-regret-bound-theorem-3-5",
        "title": "Claim 2: Regret bound (Theorem 3.5)",
        "file": "pages/claim-2-regret-bound-theorem-3-5/page.md",
        "children": []
      },
      {
        "slug": "claim-3-efficiency-separation-theorem-3-7",
        "title": "Claim 3: Efficiency separation (Theorem 3.7)",
        "file": "pages/claim-3-efficiency-separation-theorem-3-7/page.md",
        "children": []
      },
      {
        "slug": "claim-4-matgame-8-agent-10-action-nonlinear",
        "title": "Claim 4: MatGame 8-agent 10-action nonlinear",
        "file": "pages/claim-4-matgame-8-agent-10-action-nonlinear/page.md",
        "children": []
      },
      {
        "slug": "claim-5-smac-win-rate",
        "title": "Claim 5: SMAC win rate",
        "file": "pages/claim-5-smac-win-rate/page.md",
        "children": []
      },
      {
        "slug": "claim-6-smacv2-win-rate-ablations",
        "title": "Claim 6: SMACv2 win rate + ablations",
        "file": "pages/claim-6-smacv2-win-rate-ablations/page.md",
        "children": []
      },
      {
        "slug": "conclusion",
        "title": "Conclusion",
        "file": "pages/conclusion/page.md",
        "children": []
      },
      {
        "slug": "claim-1-diffusion-models-trained-on-non-overlapping-data-splits-produce-similar-outputs-for-the-same-noise-seed-and-this-is-already-largely-explained-by-a-linear-gaussian-model",
        "title": "Claim 1: Diffusion models trained on non-overlapping data splits produce similar outputs for the same noise seed, and this is already largely explained by a linear Gaussian model.",
        "file": "pages/claim-1-diffusion-models-trained-on-non-overlapping-data-splits-produce-similar-outputs-for-the-same-noise-seed-and-this-is-already-largely-explained-by-a-linear-gaussian-model/page.md",
        "children": []
      },
      {
        "slug": "executive-summary",
        "title": "Executive summary",
        "file": "pages/executive-summary/page.md",
        "children": []
      }
    ]
  },
  "traces": [],
  "workspace": {
    "file": "workspace.json",
    "file_count": 0,
    "total_size": 0,
    "bucket_id": null
  },
  "agent_view_tokens": 4577,
  "trace_view_tokens": 10,
  "workspace_view_tokens": 8,
  "revision": "bed7308d9f7a52b29211",
  "workspace_ref": {
    "repo_id": "byte-vortex/nonzero-repro-artifacts",
    "repo_type": "bucket",
    "repo_url": "https://huggingface.co/buckets/byte-vortex/nonzero-repro-artifacts",
    "private": true
  },
  "workspace_bucket": "https://huggingface.co/buckets/byte-vortex/nonzero-repro-artifacts"
}