Pith Number

pith:F4G3NPNQ

pith:2026:F4G3NPNQHBDGIS4EIBY265E52O

not attested not anchored not stored refs resolved

Children's English Reading Story Generation via Supervised Fine-Tuning of Compact LLMs with Controllable Difficulty and Safety

Bonnie J. Dorr (1), Fanghua Cao (1), Gainesville, Min Yao (1), Qian Shen (1), Shlok Gilda (1), USA), Walter L. Leite (1) ((1) University of Florida

Fine-tuned 8B LLMs generate children's reading stories that better match target difficulty levels than zero-shot outputs from GPT-4o or Llama 3.3 70B.

arxiv:2605.13709 v1 · 2026-05-13 · cs.CL · cs.AI · cs.LG

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{F4G3NPNQHBDGIS4EIBY265E52O}

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 evaluation results show that with appropriate fine-tuning designs, children's English reading stories generated by 8B LLMs perform better on difficulty-related metrics than those from zero-shot GPT-4o and Llama 3.3 70B, with almost no discernible safety issues.

C2weakest assumption

That the chosen quantitative difficulty metrics and qualitative safety checks accurately capture real-world readability and safety for children, and that fine-tuning on GPT-generated stories does not introduce hidden biases or reduce engagement.

C3one line summary

Fine-tuned 8B LLMs produce children's English reading stories with better difficulty control and safety than zero-shot GPT-4o and Llama 3.3 70B.

References

49 extracted · 49 resolved · 2 Pith anchors

[1] Journal of Children and Media , volume= 2025

[2] Reading and Writing , volume= 2006

[3] Journal of Research in Reading , volume= 2015

[4] Procedia-Social and Behavioral Sciences , volume= 2014

[5] Generative AI and ChatGPT in school children’s education: Evidence from a school lesson , author=. Sustainability , volume=. 2023 , publisher= 2023

Formal links

2 machine-checked theorem links

Receipt and verification

First computed	2026-05-18T02:44:16.774397Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

2f0db6bdb03846644b844071af749dd387d5795f4709ad579b6ffde97a6ca50c

Aliases

arxiv: 2605.13709 · arxiv_version: 2605.13709v1 · doi: 10.48550/arxiv.2605.13709 · pith_short_12: F4G3NPNQHBDG · pith_short_16: F4G3NPNQHBDGIS4E · pith_short_8: F4G3NPNQ

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/F4G3NPNQHBDGIS4EIBY265E52O \
  | 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: 2f0db6bdb03846644b844071af749dd387d5795f4709ad579b6ffde97a6ca50c

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "eb1a0dcf05cd4e12868a86f720c5105972940b3006ffe4a25e9e7463c09b77c9",
    "cross_cats_sorted": [
      "cs.AI",
      "cs.LG"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.CL",
    "submitted_at": "2026-05-13T15:56:37Z",
    "title_canon_sha256": "4eabab187baddfd8d928f3847a3196b5331f46c48e2c251014d5649f4e0a31fe"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.13709",
    "kind": "arxiv",
    "version": 1
  }
}