Buck Shlegeris

According to our database1, Buck Shlegeris authored at least 16 papers between 2018 and 2024.

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

Timeline

2018
2019
2020
2021
2022
2023
2024
0
5
10
9
2
1
1
1
1
1

Legend:

Book 
In proceedings 
Article 
PhD thesis 
Dataset
Other 

Links

On csauthors.net:

Bibliography

2024
Language Models Are Better Than Humans at Next-token Prediction.
Trans. Mach. Learn. Res., 2024

Alignment faking in large language models.
CoRR, 2024

Subversion Strategy Eval: Evaluating AI's stateless strategic capabilities against control protocols.
CoRR, 2024

Adaptive Deployment of Untrusted LLMs Reduces Distributed Threats.
CoRR, 2024

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

Sabotage Evaluations for Frontier Models.
CoRR, 2024

Games for AI Control: Models of Safety Evaluations of AI Deployment Protocols.
CoRR, 2024

Sycophancy to Subterfuge: Investigating Reward-Tampering in Large Language Models.
CoRR, 2024

Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training.
CoRR, 2024

AI Control: Improving Safety Despite Intentional Subversion.
Proceedings of the Forty-first International Conference on Machine Learning, 2024

2023
Generalized Wick Decompositions.
CoRR, 2023

Measurement Tampering Detection Benchmark.
CoRR, 2023

Interpretability in the Wild: a Circuit for Indirect Object Identification in GPT-2 Small.
Proceedings of the Eleventh International Conference on Learning Representations, 2023

2022
Polysemanticity and Capacity in Neural Networks.
CoRR, 2022

Adversarial training for high-stakes reliability.
Proceedings of the Advances in Neural Information Processing Systems 35: Annual Conference on Neural Information Processing Systems 2022, 2022

2018
Supervising strong learners by amplifying weak experts.
CoRR, 2018


  Loading...