Pith Number

pith:VOWDRF6X

pith:2026:VOWDRF6XRAOWX3FIFQE4A4XW5Q

not attested not anchored not stored refs resolved

D-VLA: A High-Concurrency Distributed Asynchronous Reinforcement Learning Framework for Vision-Language-Action Models

Haodong Yue, Haoran Sun, Junwu Xiong, Luqiao Wang, Shuai Di, Wen Huang, Xiaolong Xiang, Yicheng Gong, Yongjian Guo, Yucheng Guo, Zhen Sun, Zhong Guan

D-VLA decouples simulation and optimization planes to deliver linear speedup for trillion-parameter vision-language-action models in distributed RL.

arxiv:2605.13276 v2 · 2026-05-13 · cs.AI · cs.RO

Open paper page JSON Open Graph Bundle Merged state 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

✓ 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

Experiments on benchmarks like LIBERO show that D-VLA significantly outperforms mainstream RL frameworks in throughput and sampling efficiency for billion-parameter VLA models. In trillion-parameter scalability tests, our framework maintains exceptional stability and linear speedup.

C2weakest assumption

The assumption that physically isolating high-frequency training data from low-frequency weight control via Plane Decoupling, combined with the swimlane pipeline, eliminates interference and achieves linear speedup without new bottlenecks in real distributed hardware.

C3one line summary

D-VLA uses plane decoupling and a swimlane pipeline to deliver higher throughput and linear speedup than prior RL frameworks when training billion- and trillion-parameter VLA models on benchmarks like LIBERO.

References

28 extracted · 28 resolved · 13 Pith anchors

[1] Sanketi, Grecia Salazar, Michael S 2023

[2] $\pi_0$: A Vision-Language-Action Flow Model for General Robot Control 2024 · arXiv:2410.24164

[3] Gemini Robotics 1.5: Pushing the Frontier of Generalist Robots with Advanced Embodied Reasoning, Thinking, and Motion Transfer 2025 · arXiv:2510.03342

[4] OpenVLA: An Open-Source Vision-Language-Action Model 2024 · arXiv:2406.09246

[5] Smolvla: A vision-language-action model for affordable and efficient robotics, 2025 2025

Formal links

2 machine-checked theorem links

Receipt and verification

First computed	2026-05-18T02:44:49.236278Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

abac3897d7881d6beca82c09c072f6ec37b8fe170daca1cb26bfb28f3a8b1e1b

Aliases

arxiv: 2605.13276 · arxiv_version: 2605.13276v2 · doi: 10.48550/arxiv.2605.13276 · pith_short_12: VOWDRF6XRAOW · pith_short_16: VOWDRF6XRAOWX3FI · pith_short_8: VOWDRF6X

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/VOWDRF6XRAOWX3FIFQE4A4XW5Q \
  | 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: abac3897d7881d6beca82c09c072f6ec37b8fe170daca1cb26bfb28f3a8b1e1b

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "30d65e77cdcf7bba6dd8d9414bea4b24b3463382a50313501e9aca6ca83013df",
    "cross_cats_sorted": [
      "cs.RO"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.AI",
    "submitted_at": "2026-05-13T09:54:31Z",
    "title_canon_sha256": "e67c7c00cd97c2d5db41021910748ae0e91d5ee597a790f73cfb8c8cd9674275"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.13276",
    "kind": "arxiv",
    "version": 2
  }
}