pith:FDXUCHFT
AViS-Mamba: Adaptive Visual Steering of Audio State-Space Dynamics for Violence Detection
Video CLS tokens steer audio Mamba parameters via conditional LoRA to improve multimodal violence detection accuracy.
arxiv:2604.03329 v2 · 2026-04-02 · cs.CV · cs.AI · cs.LG · cs.SD
Add to your LaTeX paper
\usepackage{pith}
\pithnumber{FDXUCHFTMJF4H7XRHKYNLNHYA5}
Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge
Record completeness
Claims
On audio-filtered clip-level subsets of NTU-CCTV and DVD, CoLoRSMamba outperforms representative audio-only, video-only, and multimodal baselines, achieving 88.63% accuracy / 86.24% F1-V on NTU-CCTV and 75.77% accuracy / 72.94% F1-V on DVD while offering a favorable accuracy-efficiency tradeoff with fewer parameters and FLOPs.
That the CLS token from VideoMamba can reliably generate channel-wise modulation vectors and stabilization gates that adapt AudioMamba's Delta, B, C parameters to produce genuinely scene-aware audio dynamics without introducing misalignment or noise in real-world conditions.
CoLoRSMamba steers AudioMamba using video CLS-guided conditional LoRA to adapt selective state-space parameters, outperforming baselines on audio-filtered NTU-CCTV and DVD subsets with 88.63% and 75.77% accuracy respectively.
Receipt and verification
| First computed | 2026-07-07T02:18:39.570096Z |
|---|---|
| Builder | pith-number-builder-2026-05-17-v1 |
| Signature | Pith Ed25519
(pith-v1-2026-05) · public key |
| Schema | pith-number/v1.0 |
Canonical hash
28ef411cb3624bc3fef13ab0d5b4f8074249d28929eb08f580251bd43ce8ed0b
Aliases
· · · · ·Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/FDXUCHFTMJF4H7XRHKYNLNHYA5 \
| 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: 28ef411cb3624bc3fef13ab0d5b4f8074249d28929eb08f580251bd43ce8ed0b
Canonical record JSON
{
"metadata": {
"abstract_canon_sha256": "92c57b3adfce891048610ac17661527dfcd3a71c3cb7e0c30b48c00e3a3b5a25",
"cross_cats_sorted": [
"cs.AI",
"cs.LG",
"cs.SD"
],
"license": "http://creativecommons.org/licenses/by/4.0/",
"primary_cat": "cs.CV",
"submitted_at": "2026-04-02T22:14:25Z",
"title_canon_sha256": "e8cab262b431ef21758c7550d53c6438f5b737844fd2fcdfb66a1b470757e399"
},
"schema_version": "1.0",
"source": {
"id": "2604.03329",
"kind": "arxiv",
"version": 2
}
}