pith. sign in
Pith Number

pith:DMNDIYZ6

pith:2026:DMNDIYZ6LD3KV44VFERAOOUZQY
not attested not anchored not stored refs pending

From Pixels to BFS: High Maze Accuracy Does Not Imply Visual Planning

Alberto G. Rodriguez Salgado

Multimodal models solve maze images by converting them to text grids and enumerating paths token by token rather than through visual planning.

arxiv:2603.26839 v2 · 2026-03-27 · cs.LG · cs.CV

Add to your LaTeX paper
\usepackage{pith}
\pithnumber{DMNDIYZ6LD3KV44VFERAOOUZQY}

Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge

Record completeness

1 Bitcoin timestamp
2 Internet Archive
3 Author claim open · sign in to claim
4 Citations open
5 Replications open
Portable graph bundle live · download bundle · merged state
The bundle contains the canonical record plus signed events. A mirror can host it anywhere and recompute the same current state with the deterministic merge algorithm.

Claims

C1strongest claim

MazeBench therefore shows that high accuracy on visual planning tasks does not imply human-like spatial understanding.

C2weakest assumption

That the two-stage image-to-grid plus token enumeration strategy observed in traces is the dominant mechanism driving performance rather than an artifact of the specific prompting or model configurations tested.

C3one line summary

Multimodal models achieve high maze accuracy by translating images to text grids and performing token-level BFS search, not through visual planning.

Formal links

2 machine-checked theorem links

Receipt and verification
First computed 2026-05-18T03:09:22.499408Z
Builder pith-number-builder-2026-05-17-v1
Signature Pith Ed25519 (pith-v1-2026-05) · public key
Schema pith-number/v1.0

Canonical hash

1b1a34633e58f6aaf3952922073a99861b264bfad4e1494fc56067e07e3f0740

Aliases

arxiv: 2603.26839 · arxiv_version: 2603.26839v2 · doi: 10.48550/arxiv.2603.26839 · pith_short_12: DMNDIYZ6LD3K · pith_short_16: DMNDIYZ6LD3KV44V · pith_short_8: DMNDIYZ6
Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/DMNDIYZ6LD3KV44VFERAOOUZQY \
  | jq -c '.canonical_record' \
  | python3 -c "import sys,json,hashlib; b=json.dumps(json.loads(sys.stdin.read()), sort_keys=True, separators=(',',':'), ensure_ascii=False).encode(); print(hashlib.sha256(b).hexdigest())"
# expect: 1b1a34633e58f6aaf3952922073a99861b264bfad4e1494fc56067e07e3f0740
Canonical record JSON
{
  "metadata": {
    "abstract_canon_sha256": "7111c05fde22dd9f75c971e49ca175237b4901d279b78cd95c37ba1faa83da92",
    "cross_cats_sorted": [
      "cs.CV"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.LG",
    "submitted_at": "2026-03-27T08:10:05Z",
    "title_canon_sha256": "643eafe4d72694bb367d0827179831d9525fd75bafb8ce520b2208c13252d20c"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2603.26839",
    "kind": "arxiv",
    "version": 2
  }
}