Read original ↗
paperarXivTrust 82 · PrimaryPublished 11d agoLive · 10d ago

Beyond the Transcript: Detecting Covert Co ordination in Latent Multi-Agent Communication

Language-model agents can communicate through continuous hidden states that are invisible in public transcripts, creating opportunities for covert harmful coordination. We introduce Verifiable Latent Alignments (VLA), an activation-aware framework for monitoring and steering these private communication channels. For every monitored decision, VLA links the private latent-state record and channel status to the resulting public action using a shared event identifier, enabling matched causal analysis. Our first contribution is a neutral-only three-layer monitor combining representation anomaly det

Lineage graph

Paper → model → repo connections mined from source citations (Tier-1 exact match).

Why these links exist

Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.

  • FuzzySimilar title/name (fuzzy) · 59%AgentCore-8B

    Fuzzy title match (0.73): “Beyond the Transcript: Detecting Covert Co ordination in Lat” ≈ “AgentCore-8B”

  • FuzzySimilar title/name (fuzzy) · 87%SWE-agent/SWE-agent

    Fuzzy title match (0.94): “Beyond the Transcript: Detecting Covert Co ordination in Lat” ≈ “SWE-agent/SWE-agent”

  • FuzzySimilar title/name (fuzzy) · 87%zhayujie/CowAgent

    Fuzzy title match (0.94): “Beyond the Transcript: Detecting Covert Co ordination in Lat” ≈ “zhayujie/CowAgent”

  • FuzzySimilar title/name (fuzzy) · 66%open-multi-agent/open-multi-agent

    Fuzzy title match (0.78): “Beyond the Transcript: Detecting Covert Co ordination in Lat” ≈ “open-multi-agent/open-multi-agent”

  • FuzzySimilar title/name (fuzzy) · 59%NousResearch/hermes-agent

    Fuzzy title match (0.73): “Beyond the Transcript: Detecting Covert Co ordination in Lat” ≈ “NousResearch/hermes-agent”

  • FuzzySimilar title/name (fuzzy) · 59%2FastLabs/agent-squad

    Fuzzy title match (0.73): “Beyond the Transcript: Detecting Covert Co ordination in Lat” ≈ “2FastLabs/agent-squad”

  • LinkedLinked via arxiv author · 85%Ramneet Kaur

    Beyond the Transcript: Detecting Covert Co ordination in Latent Multi-Agent Communication

Covers

Has model

Implements (incoming)

authored (incoming)

Related across the graph

Topics