Torthaí cuardaigh - Leike, Jan

Hidden Incentives for Auto-Induced Distributional Shift de réir Krueger, David, Maharaj, Tegan, Leike, Jan

Faigh an téacs iomlán
Faigh an téacs iomlán

Pitfalls of learning a reward function online de réir Armstrong, Stuart, Leike, Jan, Orseau, Laurent, Legg, Shane

Faigh an téacs iomlán
Faigh an téacs iomlán

Active Reinforcement Learning: Observing Rewards at a Cost de réir Krueger, David, Leike, Jan, Evans, Owain, Salvatier, John

Faigh an téacs iomlán
Faigh an téacs iomlán

Limit-Computable Grains of Truth for Arbitrary Computable Extensive-Form (Un)Known Games de réir Wyeth, Cole, Hutter, Marcus, Leike, Jan, Taylor, Jessica

Faigh an téacs iomlán
Faigh an téacs iomlán

Excess Description Length of Learning Generalizable Predictors de réir Donoway, Elizabeth, Joren, Hailey, Roger, Fabien, Leike, Jan

Faigh an téacs iomlán
Faigh an téacs iomlán

Quantifying Differences in Reward Functions de réir Gleave, Adam, Dennis, Michael, Legg, Shane, Russell, Stuart, Leike, Jan

Faigh an téacs iomlán
Faigh an téacs iomlán

Learning Human Objectives by Evaluating Hypothetical Behavior de réir Reddy, Siddharth, Dragan, Anca D., Levine, Sergey, Legg, Shane, Leike, Jan

Faigh an téacs iomlán
Faigh an téacs iomlán

Institutionalising Ethics in AI through Broader Impact Requirements de réir Prunkl, Carina, Ashurst, Carolyn, Anderljung, Markus, Webb, Helena, Leike, Jan, Dafoe, Allan

Faigh an téacs iomlán
Faigh an téacs iomlán

Safe Deep RL in 3D Environments using Human Feedback de réir Rahtz, Matthew, Varma, Vikrant, Kumar, Ramana, Kenton, Zachary, Legg, Shane, Leike, Jan

Faigh an téacs iomlán
Faigh an téacs iomlán

Deep reinforcement learning from human preferences de réir Christiano, Paul, Leike, Jan, Brown, Tom B., Martic, Miljan, Legg, Shane, Amodei, Dario

Faigh an téacs iomlán
Faigh an téacs iomlán

Self-critiquing models for assisting human evaluators de réir Saunders, William, Yeh, Catherine, Wu, Jeff, Bills, Steven, Ouyang, Long, Ward, Jonathan, Leike, Jan

Faigh an téacs iomlán
Faigh an téacs iomlán

Prover-Verifier Games improve legibility of LLM outputs de réir Kirchner, Jan Hendrik, Chen, Yining, Edwards, Harri, Leike, Jan, McAleese, Nat, Burda, Yuri

Faigh an téacs iomlán
Faigh an téacs iomlán

Recursively Summarizing Books with Human Feedback de réir Wu, Jeff, Ouyang, Long, Ziegler, Daniel M., Stiennon, Nisan, Lowe, Ryan, Leike, Jan, Christiano, Paul

Faigh an téacs iomlán
Faigh an téacs iomlán

LLM Critics Help Catch LLM Bugs de réir McAleese, Nat, Pokorny, Rai Michael, Uribe, Juan Felipe Ceron, Nitishinskaya, Evgenia, Trebacz, Maja, Leike, Jan

Faigh an téacs iomlán
Faigh an téacs iomlán

Let's Verify Step by Step de réir Lightman, Hunter, Kosaraju, Vineet, Burda, Yura, Edwards, Harri, Baker, Bowen, Lee, Teddy, Leike, Jan, Schulman, John, Sutskever, Ilya, Cobbe, Karl

Faigh an téacs iomlán
Faigh an téacs iomlán

Scaling and evaluating sparse autoencoders de réir Gao, Leo, la Tour, Tom Dupré, Tillman, Henk, Goh, Gabriel, Troll, Rajan, Radford, Alec, Sutskever, Ilya, Leike, Jan, Wu, Jeffrey

Faigh an téacs iomlán
Faigh an téacs iomlán

Forecasting Rare Language Model Behaviors de réir Jones, Erik, Tong, Meg, Mu, Jesse, Mahfoud, Mohammed, Leike, Jan, Grosse, Roger, Kaplan, Jared, Fithian, William, Perez, Ethan, Sharma, Mrinank

Faigh an téacs iomlán
Faigh an téacs iomlán

Weak-to-Strong Generalization: Eliciting Strong Capabilities With Weak Supervision de réir Burns, Collin, Izmailov, Pavel, Kirchner, Jan Hendrik, Baker, Bowen, Gao, Leo, Aschenbrenner, Leopold, Chen, Yining, Ecoffet, Adrien, Joglekar, Manas, Leike, Jan, Sutskever, Ilya, Wu, Jeff

Faigh an téacs iomlán
Faigh an téacs iomlán

Unsupervised Elicitation of Language Models de réir Wen, Jiaxin, Ankner, Zachary, Somani, Arushi, Hase, Peter, Marks, Samuel, Goldman-Wetzler, Jacob, Petrini, Linda, Sleight, Henry, Burns, Collin, He, He, Feng, Shi, Perez, Ethan, Leike, Jan

Faigh an téacs iomlán
Faigh an téacs iomlán

Reasoning Models Don't Always Say What They Think de réir Chen, Yanda, Benton, Joe, Radhakrishnan, Ansh, Uesato, Jonathan, Denison, Carson, Schulman, John, Somani, Arushi, Hase, Peter, Wagner, Misha, Roger, Fabien, Mikulik, Vlad, Bowman, Samuel R., Leike, Jan, Kaplan, Jared, Perez, Ethan

Faigh an téacs iomlán
Faigh an téacs iomlán