pith. sign in
Pith Number

pith:K6JGKNLI

pith:2025:K6JGKNLIUJKGSEVPTRP3LSOXJG
not attested not anchored not stored refs pending

OpenThinkIMG: Learning to Think with Images via Visual Tool Reinforcement Learning

Guanjie Chen, Jiawei Gu, Juntao Li, Jun Zhang, Linjie Li, Mingyang Song, Xiaoye Qu, Yu Cheng, Yunzhuo Hao, Zhaochen Su, Zhengyuan Yang

Reinforcement learning on visual tool feedback lets a small LVLM learn adaptive tool-use policies that outperform supervised training and some larger models on chart reasoning.

arxiv:2505.08617 v2 · 2025-05-13 · cs.CV

Add to your LaTeX paper
\usepackage{pith}
\pithnumber{K6JGKNLIUJKGSEVPTRP3LSOXJG}

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

Our RL-trained agent, built upon a Qwen2-VL-2B, significantly outperforms its SFT-initialized counterpart (+28.83 points) and surpasses established supervised tool-learning baselines like Taco and CogCom by an average of +12.7 points. Notably, it also surpasses prominent closed-source models like GPT-4.1 by +8.68 accuracy points.

C2weakest assumption

The assumption that feedback from tool interactions on chart reasoning tasks will produce policies that generalize to other visual domains and tool sets without additional tuning or domain-specific reward shaping.

C3one line summary

OpenThinkIMG and V-ToolRL enable LVLMs to learn adaptive visual tool use via RL, yielding a Qwen2-VL-2B agent that beats its SFT version by 28.83 points and GPT-4.1 by 8.68 points on chart reasoning.

Formal links

2 machine-checked theorem links

Cited by

30 papers in Pith

Receipt and verification
First computed 2026-05-17T23:38:46.421271Z
Builder pith-number-builder-2026-05-17-v1
Signature Pith Ed25519 (pith-v1-2026-05) · public key
Schema pith-number/v1.0

Canonical hash

5792653568a2546912af9c5fb5c9d749b47e3aac31f3591d0f53730ce5221e15

Aliases

arxiv: 2505.08617 · arxiv_version: 2505.08617v2 · doi: 10.48550/arxiv.2505.08617 · pith_short_12: K6JGKNLIUJKG · pith_short_16: K6JGKNLIUJKGSEVP · pith_short_8: K6JGKNLI
Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/K6JGKNLIUJKGSEVPTRP3LSOXJG \
  | 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: 5792653568a2546912af9c5fb5c9d749b47e3aac31f3591d0f53730ce5221e15
Canonical record JSON
{
  "metadata": {
    "abstract_canon_sha256": "d3cec06cfc35f7bb27006d3c59e03cf69c4435b409847f692a8c93cf6d4e4e2c",
    "cross_cats_sorted": [],
    "license": "http://creativecommons.org/licenses/by-sa/4.0/",
    "primary_cat": "cs.CV",
    "submitted_at": "2025-05-13T14:35:51Z",
    "title_canon_sha256": "3ba2bef8bf3e1210a9ca37ae961fff081ff482833b3e6e2b4b0ceb9e86148ac2"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2505.08617",
    "kind": "arxiv",
    "version": 2
  }
}