pith. sign in
Pith Number

pith:JMGKQMSG

pith:2026:JMGKQMSGLCJKFGTIOLIWXCNN5H
not attested not anchored not stored refs resolved

Self-Improving CAD Generation Agents with Finite Element Analysis as Feedback

Guijin Son, Jehyun Park, Seyeon Park, Sunghee Ahn, Youngjae Yu

CAD agents improve designs when finite element analysis and blueprint feedback close the loop between generation and engineering checks.

arxiv:2605.17448 v1 · 2026-05-17 · cs.GR · cs.CL

Add to your LaTeX paper
\usepackage{pith}
\pithnumber{JMGKQMSGLCJKFGTIOLIWXCNN5H}

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

Codex (GPT-5.5) and Claude Code (Opus-4.7) agents do not produce a single strict-passing artifact in the main first-attempt sweep, with the best configuration meeting only about 20% of typed requirements on average. The same feedback tools improve geometric reconstruction, with GPT-5.5/xhigh rising from 0.444 to 0.592 Box-IoU on S2O and from 0.397 to 0.505 on Fusion360.

C2weakest assumption

That finite element analysis on the generated STEP files provides a reliable and sufficient proxy for real engineering requirements and that the observed IoU gains are causally due to the new blueprint and image feedback rather than other unmeasured factors in the agent loop. This premise enters in the abstract's description of the validation process and the reported improvements.

C3one line summary

CAD agents using finite element analysis feedback plus new text blueprint and multi-view image signals improve geometric accuracy on S2O and Fusion360 benchmarks while addressing physical validity gaps in prior generation methods.

References

37 extracted · 37 resolved · 3 Pith anchors

[1] gpt-oss-120b & gpt-oss-20b Model Card 2025 · arXiv:2508.10925
[2] Generating cad code with vision-language models for 3d designs 2024
[3] Developing a computer use model 2024
[4] Cadsmith: Multi-agent cad genera- tion with programmatic geometric validation.arXiv preprint arXiv:2603.26512, 2026 2026
[5] Geoffrey Boothroyd, Peter Dewhurst, and Winston A Knight.Product design for manufacture and assembly. CRC press, 2010 2010

Formal links

2 machine-checked theorem links

Receipt and verification
First computed 2026-05-20T00:04:39.500379Z
Builder pith-number-builder-2026-05-17-v1
Signature Pith Ed25519 (pith-v1-2026-05) · public key
Schema pith-number/v1.0

Canonical hash

4b0ca832465892a29a6872d16b89ade9f8d963409e66c46659361a197872bcaf

Aliases

arxiv: 2605.17448 · arxiv_version: 2605.17448v1 · doi: 10.48550/arxiv.2605.17448 · pith_short_12: JMGKQMSGLCJK · pith_short_16: JMGKQMSGLCJKFGTI · pith_short_8: JMGKQMSG
Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/JMGKQMSGLCJKFGTIOLIWXCNN5H \
  | 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: 4b0ca832465892a29a6872d16b89ade9f8d963409e66c46659361a197872bcaf
Canonical record JSON
{
  "metadata": {
    "abstract_canon_sha256": "df34a9e994c397bc27c0cf30ea9fee8bd38fec945293dcfd864559e1e12ccc7d",
    "cross_cats_sorted": [
      "cs.CL"
    ],
    "license": "http://creativecommons.org/licenses/by-sa/4.0/",
    "primary_cat": "cs.GR",
    "submitted_at": "2026-05-17T13:47:38Z",
    "title_canon_sha256": "69a58509d33aa03f38302fda1996a48ca665746c4b7b9714778dbedb72324e1a"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.17448",
    "kind": "arxiv",
    "version": 1
  }
}