Pith Number

pith:2VWFERGX

pith:2024:2VWFERGXU6AVU5WXWE4QRSVOEF

not attested not anchored not stored refs resolved

TraceVLA: Visual Trace Prompting Enhances Spatial-Temporal Awareness for Generalist Robotic Policies

Andrey Kolobov, Furong Huang, Hal Daum\'e III, Jianfeng Gao, Jianwei Yang, Ruijie Zheng, Shuaiyi Huang, Yongyuan Liang

Visual trace prompting encodes state-action trajectories to improve spatial-temporal awareness in vision-language-action robotic policies.

arxiv:2412.10345 v3 · 2024-12-13 · cs.RO · cs.AI

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{2VWFERGXU6AVU5WXWE4QRSVOEF}

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

Evaluations of TraceVLA across 137 configurations in SimplerEnv and 4 tasks on a physical WidowX robot demonstrate state-of-the-art performance, outperforming OpenVLA by 10% on SimplerEnv and 3.5x on real-robot tasks and exhibiting robust generalization across diverse embodiments and scenarios.

C2weakest assumption

That the 150K collected trajectories with visual traces are sufficiently diverse and representative so that the observed gains are not artifacts of the specific data-collection procedure or embodiment distribution.

C3one line summary

Visual trace prompting improves spatial-temporal awareness in VLA models, delivering 10% gains on SimplerEnv and 3.5x on real-robot tasks.

References

76 extracted · 76 resolved · 16 Pith anchors

[1] 8th Annual Conference on Robot Learning , year=

[2] 7th Annual Conference on Robot Learning , year=

[3] 8th Annual Conference on Robot Learning , year=

[4] Scaling Learning Algorithms Towards

[7] and Osindero, Simon and Teh, Yee Whye , journal =

Formal links

2 machine-checked theorem links

Cited by

45 papers in Pith

Large Language Models for Multi-Robot Systems: A Survey

SafeVLA: Towards Safety Alignment of Vision-Language-Action Model via Constrained Learning

DSSP: Diffusion State Space Policy with Full-History Encoding

EvoScene-VLA: Evolving Scene Beliefs Inside the Action Decoder for Chunked Robot Control

Action with Visual Primitives

Receipt and verification

First computed	2026-05-17T23:38:50.593974Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

d56c5244d7a7815a76d7b13908caae2168512f025d4eb1ade4d2cd320a194744

Aliases

arxiv: 2412.10345 · arxiv_version: 2412.10345v3 · doi: 10.48550/arxiv.2412.10345 · pith_short_12: 2VWFERGXU6AV · pith_short_16: 2VWFERGXU6AVU5WX · pith_short_8: 2VWFERGX

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/2VWFERGXU6AVU5WXWE4QRSVOEF \
  | 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: d56c5244d7a7815a76d7b13908caae2168512f025d4eb1ade4d2cd320a194744

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "dc17ed531fae1a4806ea140631ac6082ceb972ee833b35df1406fcf825b47c3d",
    "cross_cats_sorted": [
      "cs.AI"
    ],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.RO",
    "submitted_at": "2024-12-13T18:40:51Z",
    "title_canon_sha256": "af4e419f6184e6b71787583dac50fceafa88e714c0dda7e7caf84bab110c5241"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2412.10345",
    "kind": "arxiv",
    "version": 3
  }
}