Stack layer / Contrast
Harness-Layer Auto-Research Cut Agent Token Traffic 44.7-49.0% at Equal Task Performance
arXiv 2609.20519
Policy dependency
A Poisoned Validator Agent Can Stay Compromised at 100% and Still Produce 0% Unsafe Actions
arXiv 2609.17648
Stack layer
Cairn makes agent memory collective: query the community's reputation for a tool before calling it, submit evidence after
arXiv
Policy dependency / Stack layer
Agents hallucinate tools that do not exist, a 675B model does it as often as a 7B one, and merging MCP servers adds new failure surfaces
arXiv 2609.19425
Stack layer / Threat pattern
37,623 provenance-labeled agent PRs: Codex code was reverted half as often as human code, Devin's 31% more, and Claude Code PRs waited 12.6 hours for first review
arXiv 2609.17598
Policy dependency / Stack layer
A Cheap Read-Only Verifier Captures Nearly All the False-Pass Benefit of a Full Planning Stack
arXiv 2609.20474
Contrast
Half of 36 Inference Configurations Sit on the Pareto Frontier, and AWQ 4-bit Misses the Quality Floor by 5.9%
arXiv 2609.17863
Stack layer / Contrast
Infinite-Parameter LLMs generate feed-forward weights from live data instead of freezing them
arXiv