Pith Number

pith:3TM73BSX

pith:2025:3TM73BSXD4H7I5353NPVHUX6HM

not attested not anchored not stored refs resolved

A Survey on Vision-Language-Action Models: An Action Tokenization Perspective

Fengshuo Bai, Ka Nam Lui, Shaofei Cai, Shaoyang Guo, Tianrui Guan, Xiaowei Zhang, Xuchuan Huang, Yaodong Yang, Yifan Zhong, Yitao Liang, Yuanfei Wang, Yuanpei Chen, Zhang Chen, Zhiquan Qi

Vision-language-action models unify under one framework of action token chains from inputs to actions.

arxiv:2507.01925 v1 · 2025-07-02 · cs.RO

Open paper page JSON What is a Pith Number?

Record completeness

1 Bitcoin timestamp

2 Internet Archive

3 Author claim open · sign in to claim

4 Citations open

5 Replications open

Claims

C1strongest claim

current VLA models can be unified under a single framework: vision and language inputs are processed by a series of VLA modules, producing a chain of action tokens that progressively encode more grounded and actionable information, ultimately generating executable actions.

C2weakest assumption

the primary design choice distinguishing VLA models lies in how action tokens are formulated, which can be categorized into language description, code, affordance, trajectory, goal state, latent representation, raw action, and reasoning.

C3one line summary

The survey frames VLA models as pipelines that generate progressively grounded action tokens and classifies those tokens into eight types to guide future development.

References

299 extracted · 299 resolved · 58 Pith anchors

[1] On the Opportunities and Risks of Foundation Models 2021 · arXiv:2108.07258

[2] A comprehensive survey on pretrained foundation models: A history from bert to chatgpt 2024

[3] GPT-4 Technical Report 2023 · arXiv:2303.08774

[4] DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning 2025 · arXiv:2501.12948

[5] Learning transferable visual models from natural language supervision 2021

Formal links

2 machine-checked theorem links

Cited by

19 papers in Pith

DeepThinkVLA: Enhancing Reasoning Capability of Vision-Language-Action Models

GeoPredict: Leveraging Predictive Kinematics and 3D Gaussian Geometry for Precise VLA Manipulation

RL-VLA$^3$: A Flexible and Asynchronous Reinforcement Learning Framework for VLA Training

QuantVLA: Scale-Calibrated Post-Training Quantization for Vision-Language-Action Models

Red-Teaming Vision-Language-Action Models via Quality Diversity Prompt Generation for Robust Robot Policies

Receipt and verification

First computed	2026-05-17T23:38:13.882054Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

dcd9fd86571f0ff4777ddb5f53d2fe3b0b829472527471f8198f0dc3a6c6dc06

Aliases

arxiv: 2507.01925 · arxiv_version: 2507.01925v1 · doi: 10.48550/arxiv.2507.01925

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/3TM73BSXD4H7I5353NPVHUX6HM \
  | 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: dcd9fd86571f0ff4777ddb5f53d2fe3b0b829472527471f8198f0dc3a6c6dc06

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "1684ec07c21257a1da9c84eae86ef835a4a06bedfdb53bc256ef53935533bb40",
    "cross_cats_sorted": [],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.RO",
    "submitted_at": "2025-07-02T17:34:52Z",
    "title_canon_sha256": "55de44bf23e2520adab3a5805a62d82f2eff96566380ccc5cd38c9d8b684069c"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2507.01925",
    "kind": "arxiv",
    "version": 1
  }
}