Pith Number

pith:OPOV5ZMK

pith:2025:OPOV5ZMKBHVJAL765RJW3ZR6NU

not attested not anchored not stored refs resolved

R1-VL: Learning to Reason with Multimodal Large Language Models via Step-wise Group Relative Policy Optimization

Dacheng Tao, Huanjin Yao, Jiaxing Huang, Jingyi Zhang, Shijian Lu, Shunyu Liu, Xikun Zhang

Step-wise reinforcement learning enables multimodal models to improve their own reasoning beyond imitation.

arxiv:2503.12937 v2 · 2025-03-17 · cs.AI · cs.CL · cs.CV · cs.LG

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{OPOV5ZMKBHVJAL765RJW3ZR6NU}

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

With the proposed StepGRPO, we introduce R1-VL, a series of MLLMs with outstanding capabilities in step-by-step reasoning. Extensive experiments over 8 benchmarks demonstrate the superiority of our methods.

C2weakest assumption

The rule-based StepRAR and StepRVR rewards accurately identify necessary and logically sound reasoning steps without introducing bias or rewarding superficial patterns that do not reflect true understanding.

C3one line summary

R1-VL uses StepGRPO with rule-based StepRAR and StepRVR rewards to let MLLMs learn step-by-step reasoning beyond imitation of positive paths.

References

57 extracted · 57 resolved · 24 Pith anchors

[1] Claude 3.5 sonnet, 2024 2024

[2] Qwen2.5-VL Technical Report 2025 · arXiv:2502.13923

[3] Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback 2022 · arXiv:2204.05862

[4] Lan- guage models are few-shot learners 1901

[5] arXiv preprint arXiv:2406.10858 , year= 2024

Formal links

3 machine-checked theorem links

Cited by

31 papers in Pith

DepthAgent: Towards Better Universal Depth Estimation via Sample-wise Expert Selection

CurveBench: A Benchmark for Exact Topological Reasoning over Nested Jordan Curves

Reasoning Portability: Guiding Continual Learning for MLLMs in the RLVR Era

Not Every Rubric Teaches Equally: Policy-Aware Rubric Rewards for RLVR

RealSR-R1: Reinforcement Learning for Real-World Image Super-Resolution with Vision-Language Chain-of-Thought

Receipt and verification

First computed	2026-05-17T23:38:47.502673Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

73dd5ee58a09ea902ffeec536de63e6d25a25a45d4b04594ac2584b7aa0b6ef8

Aliases

arxiv: 2503.12937 · arxiv_version: 2503.12937v2 · doi: 10.48550/arxiv.2503.12937 · pith_short_12: OPOV5ZMKBHVJ · pith_short_16: OPOV5ZMKBHVJAL76 · pith_short_8: OPOV5ZMK

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/OPOV5ZMKBHVJAL765RJW3ZR6NU \
  | 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: 73dd5ee58a09ea902ffeec536de63e6d25a25a45d4b04594ac2584b7aa0b6ef8

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "2dae2665f2ff37d01907db2977b6e7d871d6e7c1dbba62aa21f45ff6f31cbf12",
    "cross_cats_sorted": [
      "cs.CL",
      "cs.CV",
      "cs.LG"
    ],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.AI",
    "submitted_at": "2025-03-17T08:51:44Z",
    "title_canon_sha256": "065382bfc88dc23099bc45466e988625a4a530f16d05101c5cff3fb7a6301b37"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2503.12937",
    "kind": "arxiv",
    "version": 2
  }
}