Pith Number

pith:ZYQFF5SA

pith:2025:ZYQFF5SARDW4SGOCW2R7236AAQ

not attested not anchored not stored refs resolved

ZeroSearch: Incentivize the Search Capability of LLMs without Searching

Fei Huang, Hao Sun, Jiayan Guo, Jingren Zhou, Pengjun Xie, Xuanbo Fan, Yan Zhang, Yingyan Hou, Yong Jiang, Zile Qiao

A fine-tuned retrieval module with degrading document quality trains LLMs to match or beat real search engines via RL without live API calls.

arxiv:2505.04588 v2 · 2025-05-07 · cs.CL

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{ZYQFF5SARDW4SGOCW2R7236AAQ}

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

a 7B retrieval module achieves comparable performance to the real search engine, while a 14B retrieval module even surpasses it.

C2weakest assumption

That progressively degrading the quality of documents generated by the fine-tuned retrieval module during curriculum rollouts will reliably elicit and improve the main model's reasoning ability in a manner that transfers to real search engine use.

C3one line summary

ZeroSearch simulates search engine interactions via supervised fine-tuning of a retrieval module and curriculum-based RL degradation of document quality, achieving comparable or superior performance to real search engines with 7B and 14B modules.

References

49 extracted · 49 resolved · 17 Pith anchors

[1] A. Asai, Z. Wu, Y . Wang, A. Sil, and H. Hajishirzi. Self-rag: Learning to retrieve, generate, and critique through self-reflection. In The Twelfth International Conference on Learning Representations 2023

[2] https://arxiv.org/abs/2212.08037 2022

[3] PaLM: Scaling Language Modeling with Pathways 2022 · arXiv:2204.02311

[4] The Llama 3 Herd of Models 2024 · arXiv:2407.21783

[5] W. Feng, C. Hao, Y . Zhang, J. Song, and H. Wang. Airrag: Activating intrinsic reasoning for retrieval augmented generation via tree-based search. arXiv preprint arXiv:2501.10053, 2025 2025

Formal links

2 machine-checked theorem links

Cited by

26 papers in Pith

Advancing Multi-Agent RAG Systems with Minimalist Reinforcement Learning

Search-E1: Self-Distillation Drives Self-Evolution in Search-Augmented Reasoning

Harnessing LLM Agents with Skill Programs

AIPO: Learning to Reason from Active Interaction

The Landscape of Agentic Reinforcement Learning for LLMs: A Survey

Receipt and verification

First computed	2026-05-17T23:38:13.492969Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

ce2052f64088edc919c2b6a3fd6fc00436fd02c0c7b01a2e6d496d72e771c351

Aliases

arxiv: 2505.04588 · arxiv_version: 2505.04588v2 · doi: 10.48550/arxiv.2505.04588 · pith_short_12: ZYQFF5SARDW4 · pith_short_16: ZYQFF5SARDW4SGOC · pith_short_8: ZYQFF5SA

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/ZYQFF5SARDW4SGOCW2R7236AAQ \
  | 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: ce2052f64088edc919c2b6a3fd6fc00436fd02c0c7b01a2e6d496d72e771c351

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "24ee6e81d6ad32332790766812db89532d98c9ec58455280211a714f2719674f",
    "cross_cats_sorted": [],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.CL",
    "submitted_at": "2025-05-07T17:30:22Z",
    "title_canon_sha256": "da79235b764bba152c0a246fa0c89a521fcd544162d904d39cd880bd9dfefac2"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2505.04588",
    "kind": "arxiv",
    "version": 2
  }
}