Pith Number

pith:SXAS6Z57

pith:2026:SXAS6Z57UKFY2ELICH6QA4EYH3

not attested not anchored not stored refs resolved

Learning Structured Robot Policies from Vision-Language Models via Synthetic Neuro-Symbolic Supervision

Alessandro Adami, Marco Todescato, Pietro Falco, Ruggero Carli, Tommaso Tubaldo

A 12B-parameter model learns to output executable Behavior Tree policies for robots from vision and language using only synthetic data.

arxiv:2604.02812 v2 · 2026-04-03 · cs.RO

Open paper page JSON Open Graph Bundle Merged state Verified badge What is a Pith Number?

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{SXAS6Z57UKFY2ELICH6QA4EYH3}

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

we demonstrate that a 12B-parameter model can learn structured spatial-symbolic mappings required for executable BT synthesis, solely through in-silico supervision. Real-world physical experiments on two heterogeneous robotic manipulators confirm that these structurally constrained policies achieve zero-shot transfer to real-world environments.

C2weakest assumption

The automated pipeline that generates a synthetic multimodal dataset of domain-randomized scenes paired with instruction-policy examples produced by a foundation model provides high-fidelity supervision sufficient for zero-shot transfer to physical robot environments.

C3one line summary

A 12B-parameter VLM learns to synthesize executable Behavior Tree policies from multimodal inputs via synthetic neuro-symbolic supervision, achieving zero-shot real-world transfer on robotic manipulators.

References

28 extracted · 28 resolved · 5 Pith anchors

[1] A survey of behavior trees in robotics and ai 2022

[2] Behavior trees in robotics and ai: An introduction 2018

[3] Vlm-driven behavior tree for context-aware task planning 2025

[4] Real2sim based on active perception with automatically vlm-generated behavior trees

[5] Available: https://arxiv.org/abs/2601.08454

Receipt and verification

First computed	2026-05-20T00:01:40.782198Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

95c12f67bfa28b8d116811fd0070983eedc17624a123521990a4519aef58bcb9

Aliases

arxiv: 2604.02812 · arxiv_version: 2604.02812v2 · doi: 10.48550/arxiv.2604.02812 · pith_short_12: SXAS6Z57UKFY · pith_short_16: SXAS6Z57UKFY2ELI · pith_short_8: SXAS6Z57

Agent API

Resolver JSON Graph JSON Events JSON Schema Signing key

Verify this Pith Number yourself

curl -sH 'Accept: application/ld+json' https://pith.science/pith/SXAS6Z57UKFY2ELICH6QA4EYH3 \
  | 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: 95c12f67bfa28b8d116811fd0070983eedc17624a123521990a4519aef58bcb9

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "9b52a2f6ce56b3aa82c69115de5f1311bfc8dfa1e315d7e085a885209cc7939c",
    "cross_cats_sorted": [],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.RO",
    "submitted_at": "2026-04-03T07:27:33Z",
    "title_canon_sha256": "17e12e9480641d6485c5544de1e8c3b5321d9e956eda821286ebc1d66dd6dfe9"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2604.02812",
    "kind": "arxiv",
    "version": 2
  }
}