pith. sign in
Pith Number

pith:JRPOBI7F

pith:2026:JRPOBI7FVFKUY4SAY5JRSPNQ6E
not attested not anchored not stored refs resolved

Large Language Models for Web Accessibility: A Systematic Literature Review

Rubel Hassan Mollik, Wajdi Aljedaani

A review of 38 studies finds LLMs mainly applied to text-centric and structural web accessibility tasks using WCAG as the core framework.

arxiv:2605.13873 v1 · 2026-05-06 · cs.DL · cs.AI · cs.HC

Add to your LaTeX paper
\usepackage{pith}
\pithnumber{JRPOBI7FVFKUY4SAY5JRSPNQ6E}

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 findings show that most studies apply LLMs to text-centric and structurally explicit accessibility tasks, with WCAG serving as the primary reference framework and limited consideration of cognitive accessibility guidelines (COGA).

C2weakest assumption

The search strategy and inclusion criteria captured a representative sample of all relevant peer-reviewed work on LLMs for web accessibility without significant publication bias or missed studies.

C3one line summary

A review of 38 studies finds LLMs mostly target text-based accessibility tasks under WCAG guidelines, with limited attention to cognitive issues and rare direct involvement of disabled users in evaluations.

References

64 extracted · 64 resolved · 0 Pith anchors

[1] W3C Working Group Note 2021
[2] Iyad Abu Doush and Reem Kassem. 2025. Can generative AI create accessible web code? A benchmark analysis of AI-generated HTML against accessibility standards.Universal Access in the Information Societ 2025
[3] Patricia Acosta-Vargas, Gloria Acosta-Vargas, Belén Salvador-Acosta, and Janio Jadán-Guerrero. 2024. Addressing web accessibility challenges with generative artificial intelligence tools for inclusive 2024
[4] CP Afsal and KS Kuppusamy. 2024. Comparative Assessment of Accessibility and Readability in Generative AI: A Case Study of GPT and Google BARD. InInternational Conference on Emerging Trends and Techno 2024
[5] CP Afsal and KS Kuppusamy. 2025. WEBSumm: A Chrome Extension for Sum- marizing Web Content Using LLMs for Visually Impaired Users.SN Computer Science6, 2 (2025), 1–15 2025

Formal links

2 machine-checked theorem links

Receipt and verification
First computed 2026-05-17T23:39:19.292523Z
Builder pith-number-builder-2026-05-17-v1
Signature Pith Ed25519 (pith-v1-2026-05) · public key
Schema pith-number/v1.0

Canonical hash

4c5ee0a3e5a9554c7240c753193db0f10b4258c840dadb8a6b8c4239a096f230

Aliases

arxiv: 2605.13873 · arxiv_version: 2605.13873v1 · doi: 10.48550/arxiv.2605.13873 · pith_short_12: JRPOBI7FVFKU · pith_short_16: JRPOBI7FVFKUY4SA · pith_short_8: JRPOBI7F
Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/JRPOBI7FVFKUY4SAY5JRSPNQ6E \
  | 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: 4c5ee0a3e5a9554c7240c753193db0f10b4258c840dadb8a6b8c4239a096f230
Canonical record JSON
{
  "metadata": {
    "abstract_canon_sha256": "4e19757030e88953080f190c04833a3cb8bc54148bc92678b9661aca5b57a97a",
    "cross_cats_sorted": [
      "cs.AI",
      "cs.HC"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.DL",
    "submitted_at": "2026-05-06T01:40:27Z",
    "title_canon_sha256": "d6c5fc98b2f92d2e4dc183bacf513a4dbbcd2a75eb170dac15999f6d4a7c3f07"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.13873",
    "kind": "arxiv",
    "version": 1
  }
}