pith:FFZKSTMR
Chain of Evidence: Pixel-Level Visual Attribution for Iterative Retrieval-Augmented Generation
Vision-language models can deliver pixel-level visual evidence chains for iterative retrieval-augmented generation by operating directly on document screenshots.
arxiv:2605.01284 v2 · 2026-05-02 · cs.CV · cs.AI · cs.CL · cs.IR
Add to your LaTeX paper
\usepackage{pith}
\pithnumber{FFZKSTMR2G5MIUBLZGEG2R6P6U}
Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge
Record completeness
Claims
fine-tuned Qwen3-VL-8B-Instruct achieves robust performance, significantly outperforming text-based baselines in scenarios requiring visual layout understanding, while establishing a retriever-agnostic solution for pixel-level interpretable iRAG.
That vision-language models applied to raw document screenshots can reliably recover spatial logic and layout cues that text conversion discards, without format-specific parsing or additional supervision beyond the fine-tuning described.
CoE applies vision-language models directly to document screenshots to deliver pixel-level bounding-box attribution for evidence in iterative retrieval-augmented generation, outperforming text baselines on visual-layout tasks.
Receipt and verification
| First computed | 2026-05-26T01:03:31.680631Z |
|---|---|
| Builder | pith-number-builder-2026-05-17-v1 |
| Signature | Pith Ed25519
(pith-v1-2026-05) · public key |
| Schema | pith-number/v1.0 |
Canonical hash
2972a94d91d1bac4502bc9886d47cff5086052b64ce46b316a36ffb3550198e7
Aliases
· · · · ·Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/FFZKSTMR2G5MIUBLZGEG2R6P6U \
| 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: 2972a94d91d1bac4502bc9886d47cff5086052b64ce46b316a36ffb3550198e7
Canonical record JSON
{
"metadata": {
"abstract_canon_sha256": "118bf059e09fbe7927df304ab6d8bb2c863f08dcaba997245484db0b223c2087",
"cross_cats_sorted": [
"cs.AI",
"cs.CL",
"cs.IR"
],
"license": "http://creativecommons.org/licenses/by/4.0/",
"primary_cat": "cs.CV",
"submitted_at": "2026-05-02T06:40:42Z",
"title_canon_sha256": "d08b1190e6ccce64ca8ad55466d40ca9103daa69a66afaeaa7dd9130256fe2f7"
},
"schema_version": "1.0",
"source": {
"id": "2605.01284",
"kind": "arxiv",
"version": 2
}
}