Skip to content
conftrace_
2026 EMNLP EMNLP 2026

Facts Without Rules: Boundary Metadata Collapse in Multi-Agent LLM Handoffs

AbstractMulti-agent LLM systems often coordinate by compressing an upstream interaction into a handoff artifact that downstream agents treat as shared state. We show that this handoff step is a structural source of privacy leakage: summaries preferentially preserve operational facts while weakening the boundary metadata that governs how those facts may be used—a failure mode we call summary collapse. On a controlled multi-agent coordination testbed we measure marker survival with a human-validated judge (𝜅 = 0.74), where 𝜎b = 1 means every boundary marker survives verbatim and 𝜎b = 0 means all are lost. Boundary-marker and operational-fact survival are nearly uncorrelated at the handoff level on both GPT-5-mini and DeepSeek-R1-32B (Pearson r near zero): uncompressed free-text handoffs preserve boundaries at 𝜎b ≈ 0.80, whereas a 25-word budget drops 𝜎b to ≈0.57 while operational-fact survival stays near ceiling. Controlled downstream tests reveal that protection depends on boundary explicitness: vague languages leak in 73% of GPT and 50% of DeepSeek cases, while explicit constraints reduce leakage to under 15% across all three tested models. A no-handoff single-agent control further shows the failure is not reducible to multi-agent topology as direct full-marker access still leaks more often than the operationalized handoff. Prompt-only mitigation and exact-string redaction only partially address the problem, while a gold-derived audience allowlist nearly eliminates leakage across models, showing that correctly identifying audience boundaries is the key factor.

More from EMNLP 2026

All EMNLP 2026 papers →