Zobrazeno 1 - 9
of 9
pro vyhledávání: '"Nadeau, Max"'
Publikováno v:
Workshop on Challenges in Deployable Generative AI at International Conference on Machine Learning (ICML), Honolulu, Hawaii, USA. 2023
Language models often exhibit behaviors that improve performance on a pre-training objective but harm performance on downstream tasks. We propose a novel approach to removing undesirable behaviors by ablating a small number of causal pathways between
Externí odkaz:
http://arxiv.org/abs/2309.05973
When training powerful AI systems to perform complex tasks, it may be challenging to provide training signals which are robust to optimization. One concern is \textit{measurement tampering}, where the AI system manipulates multiple measurements to cr
Externí odkaz:
http://arxiv.org/abs/2308.15605
Autor:
Casper, Stephen, Davies, Xander, Shi, Claudia, Gilbert, Thomas Krendl, Scheurer, Jérémy, Rando, Javier, Freedman, Rachel, Korbak, Tomasz, Lindner, David, Freire, Pedro, Wang, Tony, Marks, Samuel, Segerie, Charbel-Raphaël, Carroll, Micah, Peng, Andi, Christoffersen, Phillip, Damani, Mehul, Slocum, Stewart, Anwar, Usman, Siththaranjan, Anand, Nadeau, Max, Michaud, Eric J., Pfau, Jacob, Krasheninnikov, Dmitrii, Chen, Xin, Langosco, Lauro, Hase, Peter, Bıyık, Erdem, Dragan, Anca, Krueger, David, Sadigh, Dorsa, Hadfield-Menell, Dylan
Reinforcement learning from human feedback (RLHF) is a technique for training AI systems to align with human goals. RLHF has emerged as the central method used to finetune state-of-the-art large language models (LLMs). Despite this popularity, there
Externí odkaz:
http://arxiv.org/abs/2307.15217
Recent work has shown that computation in language models may be human-understandable, with successful efforts to localize and intervene on both single-unit features and input-output circuits. Here, we introduce an approach which extends causal media
Externí odkaz:
http://arxiv.org/abs/2307.03637
The literature on adversarial attacks in computer vision typically focuses on pixel-level perturbations. These tend to be very difficult to interpret. Recent work that manipulates the latent representations of image generators to create "feature-leve
Externí odkaz:
http://arxiv.org/abs/2110.03605
Autor:
Pouillé, Sophie1 (AUTHOR) sophie.pouille@umontreal.ca, Talbot, Julie1 (AUTHOR), Tamalavage, Anne E.2 (AUTHOR), Kessler‐Nadeau, Max Émile1 (AUTHOR), King, James1 (AUTHOR)
Publikováno v:
Journal of Geophysical Research. Biogeosciences. Jun2024, Vol. 129 Issue 6, p1-17. 17p.
Akademický článek
Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.
K zobrazení výsledku je třeba se přihlásit.
Autor:
Nadeau, Maxime
L’apparition du concept de littératie — né autour des années 1980 d’une préoccupation quant aux taux alarmants d’illettrisme des populations occidentales scolarisées, et d’un problème définitionnel concernant la notion d’illettrism
Externí odkaz:
http://savoirs.usherbrooke.ca/handle/11143/5361
Autor:
Kessler-Nadeau, Max Émile
La région de Rouyn-Noranda est fortement touchée par la contamination en éléments traces (ET), tels que l’arsenic (As), le cadmium (Cd), le cuivre (Cu) et le plomb (Pb), provenant des dépositions atmosphériques générées par les émissions
Externí odkaz:
http://hdl.handle.net/1866/26493