Codec-Robust Attacks on Audio LLMs
A new study introduces CodecAttack, an advanced method for attacking Audio Large Language Models (Audio LLMs). This technique optimizes perturbations in a neural audio codec's latent space, demonstrating significant success rates against various codecs. The findings indicate that lossy compression is not an effective defense against adversarial audio attacks.
- ▪CodecAttack achieves an average 85.5% target-substring attack success rate on Opus at moderate bitrates.
- ▪The attack transfers to held-out codecs, reaching up to 100% success on MP3 and 84% on AAC-LC without retraining.
- ▪The latent perturbation concentrates below 4kHz, where codecs allocate the most bits, while traditional waveform attacks spread into higher frequencies.
arXiv cs.AI files mainly under ai research. We currently carry 1,128 of its stories.
Story provenance
Source · retrieval · rights · ranking — open for full record
inspect →
Story provenance
Attribution is not the same as permission. This drawer separates discovery metadata, excerpts, WeSearch-generated summaries, reuse status, and whether the publisher receives the visit. Nothing here claims a legal grant the publisher has not made.
Record
| Original publisher | arXiv cs.AI |
| Canonical URL | https://arxiv.org/abs/2605.20519 |
| Publication time | Fri, 22 May 2026 00:00:00 -0400 |
| Retrieval time | 2026-05-22T04:02:00.009Z |
| Last seen | 2026-05-22T04:02:00.009Z |
| Headline source | Publisher (no WeSearch rewrite) |
| Excerpt source | publisher body |
| Excerpt method | First ~120 words (~800 chars) of extracted publisher body, fair-use limited. |
| Summary | WeSearch · cerebras-chat (WeSearch summarizer) |
| Summary source text | contentText |
| Citation coverage | Summary is a WeSearch-generated derivative; primary citation is the original publisher URL. |
| Cluster | lRhmZ66Hqrsi |
| Cluster logic | Grouped by semantic title/content similarity across sources within a rolling window. Same-publisher template collisions are excluded from coverage comparison. |
| Ranking reason | Story pages are not engagement-ranked. Hub feeds use recency, with optional source-diversified chronological ordering (cap consecutive stories per source). No personalized ranking. |
| Publisher visit | Yes — open original |
| Substitutes article? | No — link-out required for full text |
Rights status (four layers)
WeSearch handling by dimension
| Indexing | May the item be indexed (stored, ranked, made findable)? | Allowed |
| Snippet | May a short excerpt of the publisher's text be shown? | Allowed |
| AI summary | May WeSearch generate its own short summary of the article? | Limited |
| Retrieval / RAG | May the content be exposed for third-party retrieval-augmented generation? | Not asserted |
| Model training | May the content be used to train AI models? | Not asserted |
| Commercial reuse | May the content be reused commercially? | Not permitted |
Basis: Derived from the published RSS/Atom feed. Contact: [email protected]. Reviewed: 2026-07-24.
Opening excerpt (first ~120 words) tap to expand
Computer Science > Sound arXiv:2605.20519 (cs) [Submitted on 19 May 2026] Title:Codec-Robust Attacks on Audio LLMs Authors:Jaechul Roh, Jean-Philippe Monteuuis, Jonathan Petit, Amir Houmansdar View a PDF of the paper titled Codec-Robust Attacks on Audio LLMs, by Jaechul Roh and 3 other authors View PDF HTML (experimental) Abstract:Prior attacks on Audio Large Language Models (Audio LLMs) demonstrated that carefully crafted waveform-domain perturbations can force targeted adversarial outputs. As a defense mechanism against these attacks, real-world codec compression preprocessing has been studied to both detect and remove the perturbations. Yet no existing attack has demonstrated robustness against these compressions.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at arXiv cs.AI.
Discussion
0 commentsMore from arXiv cs.AI
