Shared entity / Stack layer
SpecGuard Turns Speculative Decoding's Acceptance Rate Into a Free Runtime Backdoor Detector
arXiv
Shared entity / Policy dependency
Microsoft Agent Framework .NET 1.21.0 tracks A2A task state and breaks the file-access line-numbering contract
GitHub Releases
Policy dependency / Stack layer
Holding Back Ready Agent Turns Instead of Releasing Them Eagerly Cuts P95 Workflow Latency up to 3.50x
arXiv 2609.10964
Stack layer / Threat pattern
BlueSTAR runs tiered autonomous cyber defense on two live enterprise IT/OT ranges
arXiv
Shared entity / Stack layer
Listen Labs walked from a signed $1.5B Menlo term sheet to negotiate a ~$2B Salesforce sale
TechCrunch
Policy dependency / Stack layer
Tencent study: agents cross authorization boundaries in 55-62% of runs when a control constraint is missing and an unsafe action is executable, and 87% when compaction drops the constraint
arXiv
Policy dependency / Stack layer
Mr.LHDR benchmark: the best deep-research agent gets 43.1% of final answers right but only 34.3% with a correct intermediate chain
arXiv
Stack layer / Contrast
EvoSafeHarness searches policies and code together to build a per-model safety harness, cutting attack success from 45.6% to 10.0%
arXiv (2609.05903)