Saswat Das
Saswat Das
Home
Experience
Publications
Projects
Contact
CV
Light
Dark
Automatic
3
SEABench: Benchmarking Endogenous Misalignment In Self-Evolving Agents
Saswat Das
,
Parvati Viswanathan (UVA)
,
Daniel Donnelly (ELLIS Institute Tubingen)
,
Chang Huang (UVA/UIUC)
,
Sahar Abdelnabi (ELLIS Institute Tubingen)
,
Ferdinando Fioretto (UVA)
PDF
Cite
NeuroFilter: Privacy Guardrails for Conversational LLM Agents
Saswat Das
,
Ferdinando Fioretto (UVA)
PDF
Cite
Beyond Jailbreaking: Auditing Contextual Privacy in LLM Agents
Saswat Das
,
Jameson Sandler (UVA)
,
Ferdinando Fioretto (UVA)
PDF
Cite
Cite
×