7 accepted papers
At the forefront of state-of-the-art human alignment methods are preference optimization methods (*PO). Prior research has often concentrated on identifying the best-performing method, typically involving a grid search over hyperparameters, which can be impractical for general practitioners. In this…
Achieving both high safety and high usefulness simultaneously in large language models has become a critical challenge in recent years. Models often exhibit unsafe behavior or adopt an overly cautious approach leading to frequent overrefusal of benign prompts, which reduces their usefulness. A majo…
We propose a novel scaling law for general-purpose decoder-only language models (LMs) trained on multilingual data, tackling the problem of balancing languages during multilingual pretraining. A primary challenge in studying multilingual scaling is the difficulty of analyzing individual language per…
State-of-the-art LLMs are powered by scaling -- scaling model size, training tokens, and cluster size. It is economically infeasible to extensively tune hyperparameters for the largest runs. Instead, approximately optimal hyperparameters must be inferred or transferred from smaller experiments. Hype…
One of the prominent issues stifling the current generation of large language models is their limited context length. Recent proprietary models such as GPT-4 and Claude 2 have introduced longer context lengths, 8k/32k and 100k, respectively; however, despite the efforts in the community, most common…
Position modeling plays a critical role in Transformers. In this paper, we focus on length extrapolation, i.e., training on short texts while evaluating longer sequences. We define attention resolution as an indicator of extrapolation. Then we propose two designs to improve the above metric of Trans…
A big convergence of model architectures across language, vision, speech, and multimodal is emerging. However, under the same name ''Transformers'', the above areas use different implementations for better performance, e.g., Post-LayerNorm for BERT, and Pre-LayerNorm for GPT and vision Transformers.…