Pith Number

pith:XB3B7E5E

pith:2025:XB3B7E5EZFGEUEGNGRF5LRPULS

not attested not anchored not stored refs resolved

Scaling Latent Reasoning via Looped Language Models

Andrew Smith, Bohong Wu, Boyi Wei, Chenghua Lin, Enduo Zhao, Fan Yin, Ge Zhang, Haoran Que, He Xing, Hongzhi Huang, Jason Eshraghian, Jiaheng Liu, Jiajun Shi, Jian Yang, Kai Hua, Kaijing Ma, Lu Li, Mude Hui, Qiyang Min, Rui-Jie Zhu, Shanda Li, Taylor Kergan, Tianle Cai, Tianyu Zhang, Wei Ye, Wenhao Huang, Xingwei Qu, Xun Zhou, Yoshua Bengio, Yunfeng Shi, Ziniu Li, Zixin Wen, Zixuan Wang

Looped language models match up to 12B model performance with 1.4B and 2.6B parameters by reasoning iteratively in latent space.

arxiv:2510.25741 v4 · 2025-10-29 · 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{XB3B7E5EZFGEUEGNGRF5LRPULS}

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

Ouro 1.4B and 2.6B models enjoy superior performance that match the results of up to 12B SOTA LLMs across a wide range of benchmarks. This advantage stems not from increased knowledge capacity, but from superior knowledge manipulation capabilities.

C2weakest assumption

The observed performance gains are caused by the latent iterative computation and entropy-regularized objective rather than differences in training data volume, optimization details, or other unstated architectural choices.

C3one line summary

Looped language models with latent iterative computation and entropy-regularized depth allocation achieve performance matching up to 12B standard LLMs through superior knowledge manipulation.

References

99 extracted · 99 resolved · 28 Pith anchors

[1] Language models are few-shot learners 1901

[2] Qwen2 Technical Report 2024 · arXiv:2407.10671

[3] Qwen3 Technical Report 2025 · arXiv:2505.09388

[4] Gemma 3 Technical Report 2025 · arXiv:2503.19786

[5] The llama 3 herd of models.arXiv e-prints, pages arXiv–2407 2024

Formal links

3 machine-checked theorem links

Cited by

30 papers in Pith

Training-Free Looped Transformers

LACO: Adaptive Latent Communication for Collaborative Driving

HRM-Text: Efficient Pretraining Beyond Scaling

On the Cost and Benefit of Chain of Thought: A Learning-Theoretic Perspective

Simply Stabilizing the Loop via Fully Looped Transformer

Receipt and verification

First computed	2026-05-17T23:38:53.108771Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

b8761f93a4c94c4a10cd344bd5c5f45c8c31dc1c09fa8a149b92d27727ad9912

Aliases

arxiv: 2510.25741 · arxiv_version: 2510.25741v4 · doi: 10.48550/arxiv.2510.25741 · pith_short_12: XB3B7E5EZFGE · pith_short_16: XB3B7E5EZFGEUEGN · pith_short_8: XB3B7E5E

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/XB3B7E5EZFGEUEGNGRF5LRPULS \
  | 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: b8761f93a4c94c4a10cd344bd5c5f45c8c31dc1c09fa8a149b92d27727ad9912

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "f105c0d60e4d96858f7d1c29bc54f9cce41f06ad3c95e5c75c1158ee56807267",
    "cross_cats_sorted": [],
    "license": "http://creativecommons.org/licenses/by-sa/4.0/",
    "primary_cat": "cs.CL",
    "submitted_at": "2025-10-29T17:45:42Z",
    "title_canon_sha256": "a0df220d6138fbd4b125ce65a3ca729ec8e2284bed4a8b7ef7cd2c9f9cac4707"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2510.25741",
    "kind": "arxiv",
    "version": 4
  }
}