This setup makes it possible for a detector to later identify the source of the text if the user publishes it.

However, if error is heavy-tailed, some policies obtain arbitrarily high reward despite achieving no more utility than the base model-a phenomenon we call catastrophic Goodhart. We adapt a discrete optimization method to measure the tails of reward models, finding that they are consistent with light-tailed error.

Frontier AI systems, including large language models (LLMs), hold increasing influence over the epistemology of human users. Such influence can reinforce prevailing societal values, potentially contributing to the lock-in of misguided moral beliefs and, consequently, the perpetuation of problematic moral practices on a broad scale. We introduce progress alignment as a technical solution to mitigate this imminent risk. Progress alignment algorithms learn to emulate the mechanics of human moral progress, thereby addressing the susceptibility of existing alignment methods to contemporary moral blindspots.

Training models with longer in-context lengths is a significant challenge for multi-modal machine learning due to substantial GPU memory and computational costs.

Transformers have shown impressive capabilities across various tasks, but their performance on compositional problems remains a topic of debate. In this work, we investigate the mechanisms of how transformers behave on unseen compositional tasks. We discover that the parameter initialization scale plays a critical role in determining whether the model learns inferential (reasoning-based) solutions, which capture the underlying compositional primitives, or symmetric (memory-based) solutions, which simply memorize mappings without understanding the compositional structure. By analyzing the information flow and vector representations within the model, we reveal the distinct mechanisms underlying these solution types. We further find that inferential (reasoning-based) solutions exhibit low complexity bias, which we hypothesize is a key factor enabling them to learn individual mappings for single anchors.

Bilevel optimization methods solve problems with hierarchical structures, optimizing two interdependent objectives: an inner-level objective and an outer-level one. Initially used in machine learning for model selection [Bennett et al., 2006] and sparse feature learning [Mairal et al., 2012], these methods gained popularity as efficient alternatives to grid search for hyper-parameter tuning [Feurer and Hutter,

Unlike the common practice of aligning LLMs during instruction tuning or reinforcement learning phase, referred to as ' post