Pith Number

pith:NPPY77OF

pith:2025:NPPY77OFA2MKVPOVDHAWQWJQ7L

not attested not anchored not stored refs pending

Where Do Backdoors Live? A Component-Level Analysis of Backdoor Propagation in Speech Language Models

Alexandrine Fortier, Jes\'us Villalba, Najim Dehak, Patrick Cardinal, Peter West, Thomas Thebaud

Backdoors propagate through speech language models to leave all tasks vulnerable, with their persistence depending on the targeted component.

arxiv:2510.01157 v4 · 2025-10-01 · cs.CL · cs.CR · cs.SD

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{NPPY77OFA2MKVPOVDHAWQWJQ7L}

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

Backdoors can propagate through the SLM, leaving all tasks highly vulnerable. Backdoor persistence or erasure is highly dependent on the targeted component. Poisoned samples are not directly separable from benign ones in shared multitask embeddings.

C2weakest assumption

The component analysis assumes that backdoors can be injected into and isolated within individual pipeline stages without the injection process itself creating confounding interactions across stages that would invalidate the attribution of persistence or erasure to specific components.

C3one line summary

Backdoors propagate through SLM components with persistence or erasure depending on the targeted part, and poisoned samples are not directly separable from benign ones in shared multitask embeddings.

Formal links

1 machine-checked theorem link

Cited by

1 paper in Pith

A Survey of Large Audio Language Models: Generalization, Trustworthiness, and Outlook

Receipt and verification

First computed	2026-06-11T01:09:18.209240Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

6bdf8ffdc50698aabdd519c1685930fafd88c9216a6a04507f514b2fc0021a19

Aliases

arxiv: 2510.01157 · arxiv_version: 2510.01157v4 · doi: 10.48550/arxiv.2510.01157 · pith_short_12: NPPY77OFA2MK · pith_short_16: NPPY77OFA2MKVPOV · pith_short_8: NPPY77OF

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/NPPY77OFA2MKVPOVDHAWQWJQ7L \
  | 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: 6bdf8ffdc50698aabdd519c1685930fafd88c9216a6a04507f514b2fc0021a19

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "3046a7529181a87e049ccac33fbec649aa04f72200a1bac13a3e75affd020cdd",
    "cross_cats_sorted": [
      "cs.CR",
      "cs.SD"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.CL",
    "submitted_at": "2025-10-01T17:45:04Z",
    "title_canon_sha256": "f9cf8f6ba98cefcd6366a58614bb0d40163f5c0d64c61e96839c4ae53fa04ce7"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2510.01157",
    "kind": "arxiv",
    "version": 4
  }
}