Zobrazeno 1 - 10
of 2 394
pro vyhledávání: '"Meinke, P."'
Autor:
Meinke, Alexander, Schoen, Bronson, Scheurer, Jérémy, Balesni, Mikita, Shah, Rusheb, Hobbhahn, Marius
Frontier models are increasingly trained and deployed as autonomous agent. One safety concern is that AI agents might covertly pursue misaligned goals, hiding their true capabilities and objectives - also known as scheming. We study whether models ha
Externí odkaz:
http://arxiv.org/abs/2412.04984
Autor:
Balesni, Mikita, Hobbhahn, Marius, Lindner, David, Meinke, Alexander, Korbak, Tomek, Clymer, Joshua, Shlegeris, Buck, Scheurer, Jérémy, Stix, Charlotte, Shah, Rusheb, Goldowsky-Dill, Nicholas, Braun, Dan, Chughtai, Bilal, Evans, Owain, Kokotajlo, Daniel, Bushnaq, Lucius
We sketch how developers of frontier AI systems could construct a structured rationale -- a 'safety case' -- that an AI system is unlikely to cause catastrophic outcomes through scheming. Scheming is a potential threat model where AI systems could pu
Externí odkaz:
http://arxiv.org/abs/2411.03336
Autor:
Herten, Andreas, Achilles, Sebastian, Alvarez, Damian, Badwaik, Jayesh, Behle, Eric, Bode, Mathis, Breuer, Thomas, Caviedes-Voullième, Daniel, Cherti, Mehdi, Dabah, Adel, Sayed, Salem El, Frings, Wolfgang, Gonzalez-Nicolas, Ana, Gregory, Eric B., Mood, Kaveh Haghighi, Hater, Thorsten, Jitsev, Jenia, John, Chelsea Maria, Meinke, Jan H., Meyer, Catrin I., Mezentsev, Pavel, Mirus, Jan-Oliver, Nassyr, Stepan, Penke, Carolin, Römmer, Manoel, Sinha, Ujjwal, Vieth, Benedikt von St., Stein, Olaf, Suarez, Estela, Willsch, Dennis, Zhukov, Ilya
Publikováno v:
2024 SC24: International Conference for High Performance Computing, Networking, Storage and Analysis SC
Benchmarks are essential in the design of modern HPC installations, as they define key aspects of system components. Beyond synthetic workloads, it is crucial to include real applications that represent user requirements into benchmark suites, to gua
Externí odkaz:
http://arxiv.org/abs/2408.17211
Automated Vehicle (AV) validation based on simulated testing requires unbiased evaluation and high efficiency. One effective solution is to increase the exposure to risky rare events while reweighting the probability measure. However, characterizing
Externí odkaz:
http://arxiv.org/abs/2407.07320
Autor:
Laine, Rudolf, Chughtai, Bilal, Betley, Jan, Hariharan, Kaivalya, Scheurer, Jeremy, Balesni, Mikita, Hobbhahn, Marius, Meinke, Alexander, Evans, Owain
AI assistants such as ChatGPT are trained to respond to users by saying, "I am a large language model". This raises questions. Do such models know that they are LLMs and reliably act on this knowledge? Are they aware of their current circumstances, s
Externí odkaz:
http://arxiv.org/abs/2407.04694
Autor:
Karp, Martin, Suarez, Estela, Meinke, Jan H., Andersson, Måns I., Schlatter, Philipp, Markidis, Stefano, Jansson, Niclas
The never-ending computational demand from simulations of turbulence makes computational fluid dynamics (CFD) a prime application use case for current and future exascale systems. High-order finite element methods, such as the spectral element method
Externí odkaz:
http://arxiv.org/abs/2405.05640
Autor:
Meinke, Alexander, Evans, Owain
We examine how large language models (LLMs) generalize from abstract declarative statements in their training data. As an illustration, consider an LLM that is prompted to generate weather reports for London in 2050. One possibility is that the tempe
Externí odkaz:
http://arxiv.org/abs/2312.07779
Autor:
Meinke, Jeremy, Cohen, Seth, Moore, Jenna, Böckmann, Kathrin, Mauskopf, Philip, Scannapieco, Evan
We use public data from the South Pole Telescope (SPT) and Atacama Cosmology Telescope (ACT) to measure radial profiles of the thermal Sunyaev-Zel'dovich (tSZ) effect and dust emission around massive quiescent galaxies at $z\approx1.$ Using survey da
Externí odkaz:
http://arxiv.org/abs/2306.04760
Autor:
Rottmayer, Jan, Özkaya, Emre, Satcunanathan, Sutharsan, Zhou, Beckett Y., Aehle, Max, Gauger, Nicolas R., Meinke, Matthias, Schröder, Wolfgang, Pullin, Shaun
Broadband noise reduction is a significant problem in aerospace and industrial applications. Specifically, the noise generated from the trailing edge of an airfoil poses a challenging problem with various proposed solutions. This study investigates t
Externí odkaz:
http://arxiv.org/abs/2301.13047
The transonic flow around the OAT15A airfoil is computed at buffet conditions, i.e., freestream Mach number $Ma_\infty = 0.73$, chord-based freestream Reynolds number $Re_c = 2\cdot10^6$, and angle of attack $\alpha = 3.5^\circ$ using wall-modeled LE
Externí odkaz:
http://arxiv.org/abs/2301.05889