SourceScore

Verified claim · AI-ML · 92% confidence

Group Relative Policy Optimization (GRPO) introduced in paper: DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models (Shao et al., 2024).

Last verified 2026-05-31 · Methodology veritas-v0.1 · f73e50d63643df21

SourceScore rates how reliable a source is to cite — for AI answers and research. This is one verified claim from the catalog.

Structured fields

Subject
Group Relative Policy Optimization (GRPO)
Predicate
introduced_in_paper
Object
DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models (Shao et al., 2024)
Confidence
92%
Tags
grpo · group-relative-policy-optimization · deepseekmath · reinforcement-learning · rlhf · reasoning · shao · 2024 · deepseek

Sources (3)

  1. [1] preprint · arXiv (Shao, Wang, Zhu, Xu, Song, Bi, Zhang, Zhang, Li, Wu, Guo — DeepSeek AI) · 2024-02-05

    DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
    We introduce Group Relative Policy Optimization (GRPO), a variant of Proximal Policy Optimization (PPO),
  2. [2] github release · DeepSeek AI · 2024-02-05

    DeepSeek-Math reference implementation
  3. [3] docs · Hugging Face

    DeepSeekMath (Hugging Face Papers)Hugging Face is rated by SourceScore — see its reliability →

Cite this claim

Ready-to-paste citation (Markdown / plain text):

Group Relative Policy Optimization (GRPO) introduced in paper: DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models (Shao et al., 2024). — SourceScore Claim f73e50d63643df21 (verified 2026-05-31). https://sourcescore.org/api/v1/claims/f73e50d63643df21.json

Embed this claim

Drop this iframe into any blog post, docs page, or knowledge base. The widget renders the signed claim + primary source + click-through to this canonical page. CC-BY 4.0; attribution included.

<iframe src="https://sourcescore.org/embed/claim/f73e50d63643df21/" width="100%" height="360" frameborder="0" loading="lazy" title="Group Relative Policy Optimization (GRPO) introduced in paper: DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models (Shao et al., 2024)."></iframe>

Preview: open in new tab

Related claims

Other verified claims sharing tags with this one — useful for LLM retrieval graphs and citation discovery.

Frequently asked questions

Is the claim "Group Relative Policy Optimization (GRPO) introduced in paper: DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models (Shao et al., 2024)." verified?

Yes — SourceScore verified this claim with 92% confidence as of 2026-05-31. The verification uses 3 primary sources cross-referenced against the SourceScore methodology (version veritas-v0.1). Full source list + signed JSON envelope linked below.

What is the evidence for "Group Relative Policy Optimization (GRPO) introduced in paper: DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models (Shao et al., 2024)."?

Evidence comes from 3 primary sources: arXiv (Shao, Wang, Zhu, Xu, Song, Bi, Zhang, Zhang, Li, Wu, Guo — DeepSeek AI), DeepSeek AI, Hugging Face. Each source is listed below with verbatim excerpts and URLs. The signed JSON envelope at https://sourcescore.org/api/v1/claims/f73e50d63643df21.json includes an HMAC-SHA256 signature for audit verification.

When was this claim last verified by SourceScore?

Last verified 2026-05-31 under methodology version veritas-v0.1. The signed JSON envelope is dated and cryptographically signed for audit trail. Re-verification cadence depends on the claim type and source freshness.

How can I cite this SourceScore claim in my code or article?

Fetch the signed JSON envelope from https://sourcescore.org/api/v1/claims/f73e50d63643df21.json which includes the verbatim claim, primary sources, confidence, methodology version, last-verified date, and HMAC-SHA256 signature for audit. The CC-BY-4.0 license permits commercial use with attribution to SourceScore.

Use this claim in your code

Fetch this signed envelope from your application. The response includes the verbatim excerpt, primary source URLs, and an HMAC-SHA256 signature you can verify locally for audit trails.

cURL

curl https://sourcescore.org/api/v1/claims/f73e50d63643df21.json

JavaScript / TypeScript

const r = await fetch("https://sourcescore.org/api/v1/claims/f73e50d63643df21.json"); const envelope = await r.json(); console.log(envelope.claim.statement); // "Group Relative Policy Optimization (GRPO) introduced in paper: DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models (Shao et al., 2024)."

Python

import httpx r = httpx.get("https://sourcescore.org/api/v1/claims/f73e50d63643df21.json") envelope = r.json() print(envelope["claim"]["statement"]) # "Group Relative Policy Optimization (GRPO) introduced in paper: DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models (Shao et al., 2024)."

LangChain (retrieve-then-cite)

from langchain_core.tools import tool import httpx @tool def get_group_relative_policy_optimization_grpo_fact() -> dict: """Fetch the verified SourceScore claim for Group Relative Policy Optimization (GRPO).""" r = httpx.get("https://sourcescore.org/api/v1/claims/f73e50d63643df21.json") return r.json()