Pith Number

pith:2JJKI5UQ

pith:2025:2JJKI5UQGMKXPVPXS5K573WHRC

not attested not anchored not stored refs resolved

WebThinker: Empowering Large Reasoning Models with Deep Research Capability

Guanting Dong, Hongjin Qian, Jiajie Jin, Ji-Rong Wen, Xiaoxi Li, Yongkang Wu, Yutao Zhu, Zhicheng Dou

WebThinker lets large reasoning models search the web and draft reports autonomously during reasoning.

arxiv:2504.21776 v2 · 2025-04-30 · cs.CL · cs.AI · cs.IR

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{2JJKI5UQGMKXPVPXS5K573WHRC}

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

Extensive experiments on complex reasoning benchmarks (GPQA, GAIA, WebWalkerQA, HLE) and scientific report generation tasks (Glaive) demonstrate that WebThinker significantly outperforms existing methods and strong proprietary systems.

C2weakest assumption

That the Deep Web Explorer module can reliably locate, navigate, and extract accurate information from arbitrary web pages without introducing navigation errors or factual hallucinations that propagate into the final report.

C3one line summary

WebThinker equips large reasoning models with autonomous web exploration and interleaved reasoning-drafting via a Deep Web Explorer and RL-based DPO training, yielding gains on GPQA, GAIA, and report-generation benchmarks.

References

91 extracted · 91 resolved · 23 Pith anchors

[1] Self-rag: Learn- ing to retrieve, generate, and critique through self-reflection 2024

[2] ReSearch: Learning to Reason with Search for LLMs via Reinforcement Learning 2025 · arXiv:2503.19470

[3] Towards Reasoning Era: A Survey of Long Chain-of-Thought for Reasoning Large Language Models 2025 · arXiv:2503.09567

[4] An empirical study on eliciting and improving r1-like reasoning models 2025

[5] Self-play with execution feedback: Improving instruction-following capabilities of large language models 2025

Cited by

29 papers in Pith

ZeroSearch: Incentivize the Search Capability of LLMs without Searching

Code Researcher: Deep Research Agent for Large Systems Code and Commit History

Mem-$\pi$: Adaptive Memory through Learning When and What to Generate

Automated Root-Cause Subclassification and No-Code Fix Generation for Invalid Bug Reports

Rethinking Agentic Reinforcement Learning In Large Language Models

Receipt and verification

First computed	2026-05-17T23:38:46.873797Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

d252a47690331577d5f79755dfeec78889b87740c75130c472c25ff2b5a61c87

Aliases

arxiv: 2504.21776 · arxiv_version: 2504.21776v2 · doi: 10.48550/arxiv.2504.21776 · pith_short_12: 2JJKI5UQGMKX · pith_short_16: 2JJKI5UQGMKXPVPX · pith_short_8: 2JJKI5UQ

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/2JJKI5UQGMKXPVPXS5K573WHRC \
  | 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: d252a47690331577d5f79755dfeec78889b87740c75130c472c25ff2b5a61c87

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "e99d68991c3e046a23e457994dc6f977a81a18aab14292ab79ad9980dc1b958f",
    "cross_cats_sorted": [
      "cs.AI",
      "cs.IR"
    ],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.CL",
    "submitted_at": "2025-04-30T16:25:25Z",
    "title_canon_sha256": "d518db51547776410cc9728128f4fd47506bcf25e3bad8c45c165fe41daa4a18"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2504.21776",
    "kind": "arxiv",
    "version": 2
  }
}