Mikita Balesni

According to our database1, Mikita Balesni authored at least 8 papers between 2023 and 2024.

Collaborative distances:
  • Dijkstra number2 of five.
  • Erdős number3 of four.

Timeline

2023
2024
0
1
2
3
4
5
6
7
4
2
2

Legend:

Book 
In proceedings 
Article 
PhD thesis 
Dataset
Other 

Links

On csauthors.net:

Bibliography

2024
Frontier Models are Capable of In-context Scheming.
CoRR, 2024

The Two-Hop Curse: LLMs trained on A->B, B->C fail to learn A->C.
CoRR, 2024

Towards evaluations-based safety cases for AI scheming.
CoRR, 2024

Honesty to Subterfuge: In-Context Reinforcement Learning Can Make Honest Models Reward Hack.
CoRR, 2024

Me, Myself, and AI: The Situational Awareness Dataset (SAD) for LLMs.
Proceedings of the Advances in Neural Information Processing Systems 38: Annual Conference on Neural Information Processing Systems 2024, 2024

The Reversal Curse: LLMs trained on "A is B" fail to learn "B is A".
Proceedings of the Twelfth International Conference on Learning Representations, 2024

2023
Technical Report: Large Language Models can Strategically Deceive their Users when Put Under Pressure.
CoRR, 2023

Taken out of context: On measuring situational awareness in LLMs.
CoRR, 2023


  Loading...