SIREN
Collection
SIREN is a lightweight guard model that detects harmful content from LLM internal representations. • 4 items • Updated
AI & ML interests
None defined yet.
Recent Activity
Papers
LLM Safety From Within: Detecting Harmful Content with Internal Representations
ThinkTwice: Jointly Optimizing Large Language Models for Reasoning and Self-Refinement
lilvjosephtang
updated a
collection 1 day ago
danielgmonroe
updated a
model 14 days ago
danielgmonroe
published a
model 14 days ago
lilvjosephtang
authored a
paper 17 days ago
lilvjosephtang
submitted a
paper to Daily Papers 18 days ago
ashtonanderson
updated a
model 20 days ago
danielgmonroe
updated 4
models 21 days ago
danielgmonroe
updated a
collection 22 days ago