pith.
Research
Integrity
Review
Pre-print
sign in
Physics
Mathematics
Computer Science
Biology
Finance
Statistics
Systems
Economics
pith
› author
Evan Hubinger
1
works
1
Pith-reviewed
100.0%
Recognition coverage
0
queued
works
Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training
Pith
2024 · cs.CR · verdict UNVERDICTED ·
101 Pith citing