pith. sign in
Pith Number

pith:3ZIKIPWU

pith:2026:3ZIKIPWUSCAGDA3WYQGATK72AF
not attested not anchored not stored refs resolved

When Efficiency Backfires: Cascading LLMs Trigger Cascade Failure under Adversarial Attack

Dingfan Chen, Songze Li, Zehan Sun

Adversarial attacks exploit LLM cascade designs to degrade both accuracy and efficiency.

arxiv:2605.17288 v1 · 2026-05-17 · cs.CR · cs.AI

Add to your LaTeX paper
\usepackage{pith}
\pithnumber{3ZIKIPWUSCAGDA3WYQGATK72AF}

Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more

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

We present the first study demonstrating that LLM cascade systems are susceptible to targeted adversarial manipulation, which disrupts both performance objectives and the intended cost advantages of the cascade design.

C2weakest assumption

The inclusion of lightweight front-end models and internal decision mechanisms in the cascade design necessarily expands the attack surface in ways that prior standalone-model attacks cannot exploit.

C3one line summary

LLM cascade systems are vulnerable to a new adversarial attack that simultaneously degrades accuracy and destroys the intended cost savings by targeting both the lightweight models and the escalation decision mechanism.

References

85 extracted · 85 resolved · 12 Pith anchors

[1] GPT-4 Technical Report 2023 · arXiv:2303.08774
[2] The claude 3 model family: Opus, sonnet, haiku, 2024 2024
[3] Gemini: A Family of Highly Capable Multimodal Models 2023 · arXiv:2312.11805
[4] DeepSeek-V3 Technical Report 2024 · arXiv:2412.19437
[5] Qwen Technical Report 2023 · arXiv:2309.16609
Receipt and verification
First computed 2026-05-20T00:03:50.077692Z
Builder pith-number-builder-2026-05-17-v1
Signature Pith Ed25519 (pith-v1-2026-05) · public key
Schema pith-number/v1.0

Canonical hash

de50a43ed49080618376c40c09abfa0174b4e7473684dfb4ad09dd3bf67bfad7

Aliases

arxiv: 2605.17288 · arxiv_version: 2605.17288v1 · doi: 10.48550/arxiv.2605.17288 · pith_short_12: 3ZIKIPWUSCAG · pith_short_16: 3ZIKIPWUSCAGDA3W · pith_short_8: 3ZIKIPWU
Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/3ZIKIPWUSCAGDA3WYQGATK72AF \
  | 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: de50a43ed49080618376c40c09abfa0174b4e7473684dfb4ad09dd3bf67bfad7
Canonical record JSON
{
  "metadata": {
    "abstract_canon_sha256": "77348c7a7062673abee2b46786a3e989b917ea2d9d82a22dd3c290a16b507763",
    "cross_cats_sorted": [
      "cs.AI"
    ],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.CR",
    "submitted_at": "2026-05-17T06:59:43Z",
    "title_canon_sha256": "f7eeccbc8ceb8df0ec482bbb5cb0e16463e1d0c18c7bc15110bf7e1117e49a51"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.17288",
    "kind": "arxiv",
    "version": 1
  }
}