Detecting Lies of (C)omission
Ilan Moscovitz, Nikhil Kotecha, Sam Svenningsen · Team
Submitted to Deception Detection Hackathon: Preventing AI deception. Sprint projects are early-stage work by participants, not Apart Research publications.
We introduce the concept of deceptive omission to denote deceptive non-lying behavior. We also modify a dataset and generate a second dataset to help researchers identify deception that doesn't necessarily involve lying.
Reviews
No public critique yet.
Cite this project
@misc{moscovitz2024detecting,
title = {{Detecting Lies of (C)omission}},
author = {Ilan Moscovitz and Nikhil Kotecha and Sam Svenningsen},
year = {2024},
month = jul,
note = {Submitted to Deception Detection Hackathon: Preventing AI deception, an Apart Research Sprint},
howpublished = {\url{https://apartresearch.com/sprints/projects/detecting-lies-of-(c)omission}},
url = {https://apartresearch.com/sprints/projects/detecting-lies-of-(c)omission}
}More from Deception Detection Hackathon: Preventing AI deception
- 1st place by peer reviewView project: Sandbag Detection through Model Degradation
Sandbag Detection through Model Degradation
Team Truth Serum
We propose a novel technique to detect sandbagging in LLMs by adding varying amount of noise to model weights and monitoring performance.
- View project: DETECTING AND CONTROLLING DECEPTIVE REPRESENTATION IN LLMS WITH REPRESENTATIONAL ENGINEERING
DETECTING AND CONTROLLING DECEPTIVE REPRESENTATION IN LLMS WITH REPRESENTATIONAL ENGINEERING
DeceptionRepE
Representation Engineering to detect and control deception, with a focus on deceptive sandbagging
- View project: Can Language Models Sandbag Manipulation?
Can Language Models Sandbag Manipulation?
Can Language Models Sandbag Manipulation?
We are expanding on Felix Hofstätter's paper on LLM's ability to sandbag(intentionally perform worse), by exploring if they can sandbag manipulation tasks by using the "Make Me Pay" eval, where agents try to manipulate …