Pith Number

pith:TZ5HAE6G

pith:2026:TZ5HAE6GIHVONP7KDCAAXYD3NK

not attested not anchored not stored refs resolved

Finding the Weakest Link: Adversarial Attack against Multi-Agent Communications

Claudia Szabo, Junae Kim, Maxwell Standen

Gradient-based selection of vulnerable messages and agents disrupts multi-agent reinforcement learning communications as effectively as random attacks in most cases.

arxiv:2605.13170 v1 · 2026-05-13 · cs.LG · cs.MA

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{TZ5HAE6GIHVONP7KDCAAXYD3NK}

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

Our novel message selection method achieves a similar or greater impact than random message selection across almost all tested scenarios. Our victim selection, message selection, tempo, and loss functions improve attack effectiveness in half of the thirty scenarios we tested.

C2weakest assumption

The attacker has white-box access to compute Jacobians and gradients from the victim model, which may not hold for black-box deployed systems.

C3one line summary

Jacobian-based selection of messages, agents, and timesteps combined with two new adversarial loss functions disrupts multi-agent RL communication more effectively than random perturbations in navigation, PredatorPrey, and TrafficJunction environments.

References

36 extracted · 36 resolved · 0 Pith anchors

[1] Multi-Agent Deep Reinforcement Learning Applications in Cybersecurity: Challenges and Perspectives, 2024

[2] Multi-agent reinforcement learning for cybersecurity: Approaches and challenges, 2024

[3] F. A. Oliehoek and C. Amato,A Concise Introduction to Decentralized POMDPs. Springer International Publishing, 2016 2016

[4] Learning to Communi- cate with Deep Multi-Agent Reinforcement Learning, 2016

[5] Succinct and robust multi-agent communication with temporal message control, 2020

Receipt and verification

First computed	2026-05-18T03:08:56.586105Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

9e7a7013c641eae6bfea18800be07b6a811a08f8e2d23cff9ad4e2c321cfa23d

Aliases

arxiv: 2605.13170 · arxiv_version: 2605.13170v1 · doi: 10.48550/arxiv.2605.13170 · pith_short_12: TZ5HAE6GIHVO · pith_short_16: TZ5HAE6GIHVONP7K · pith_short_8: TZ5HAE6G

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/TZ5HAE6GIHVONP7KDCAAXYD3NK \
  | 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: 9e7a7013c641eae6bfea18800be07b6a811a08f8e2d23cff9ad4e2c321cfa23d

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "ffb9ad8cbdb7f4c87f5fede5bf89636ffe422f36a7f20e9e5738f696b6656ff8",
    "cross_cats_sorted": [
      "cs.MA"
    ],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.LG",
    "submitted_at": "2026-05-13T08:32:30Z",
    "title_canon_sha256": "b0b5e0dee3ab9c6ecfb4fe894446fbdfa51fb021d51c9b47ff28006a1e332b93"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.13170",
    "kind": "arxiv",
    "version": 1
  }
}