Related papers: Language Models Grow Less Humanlike beyond Phase T…
Large pre-trained language models (LPLM) have shown spectacular success when fine-tuned on downstream supervised tasks. Yet, it is known that their performance can drastically drop when there is a distribution shift between the data used…
Recent advancements in large language models (LLMs) reveal a perplexing phenomenon in continual learning: despite extensive training, models experience significant performance declines, raising questions about task alignment and underlying…
This work presents a detailed linguistic analysis into why larger Transformer-based pre-trained language models with more parameters and lower perplexity nonetheless yield surprisal estimates that are less predictive of human reading times.…
By positing a relationship between naturalistic reading times and information-theoretic surprisal, surprisal theory (Hale, 2001; Levy, 2008) provides a natural interface between language models and psycholinguistic models. This paper…
Aligning Large Language Models (LLMs) with human preferences through finetuning is resource-intensive, motivating lightweight alternatives at test time. We address test-time alignment through the lens of sequential decision making, a…
Scaled post-training now drives many of the largest capability gains in language models (LMs), yet its effect on pretrained knowledge remains poorly understood. Not all forgetting is equal: Forgetting one fact (e.g., a U.S. president or an…
Presupposition projection in conditionals is central to theories of meaning and pragmatics, yet it remains largely unevaluated in large language models. We address this gap through a parallel behavioral study comparing human judgments and…
Self-improvement through post-training methods such as iterative preference learning has been acclaimed for enhancing the problem-solving capabilities (e.g., mathematical reasoning) of Large Language Models (LLMs) without human…
Transformer-based large language models (LLMs) have achieved strong performance across many natural language processing tasks. Nonetheless, their quadratic computational and memory requirements, particularly in self-attention layers, pose…
Research into the external behaviors and internal mechanisms of large language models (LLMs) has shown promise in addressing complex tasks in the physical world. Studies suggest that powerful LLMs, like GPT-4, are beginning to exhibit…
Pretrained language models (PTLMs) are typically learned over a large, static corpus and further fine-tuned for various downstream tasks. However, when deployed in the real world, a PTLM-based model must deal with data distributions that…
The next-token prediction (NTP) objective has been foundational in the development of modern large language models (LLMs), driving advances in fluency and generalization. However, NTP operates at the \textit{token} level, treating…
Large language models (LLMs) excel on a variety of reasoning benchmarks, but previous studies suggest they sometimes struggle to generalize to unseen questions, potentially due to over-reliance on memorized training examples. However, the…
Large Language Models (LLMs) trained on web-scale text corpora have been shown to capture world knowledge in their parameters. However, the mechanism by which language models store different types of knowledge is poorly understood. In this…
The ability to acquire latent semantics is one of the key properties that determines the performance of language models. One convenient approach to invoke this ability is to prepend metadata (e.g. URLs, domains, and styles) at the beginning…
Though there is a strong consensus that word length and frequency are the most important single-word features determining visual-orthographic access to the mental lexicon, there is less agreement as how to best capture syntactic and…
Learning dynamics, which describes how the learning of specific training examples influences the model's predictions on other examples, gives us a powerful tool for understanding the behavior of deep learning systems. We study the learning…
Large language models (LLMs) are susceptible to persuasion, which can pose risks when models are faced with an adversarial interlocutor. We take a first step towards defending models against persuasion while also arguing that defense…
Recent advancements in large language models (LLMs) have demonstrated that progressive refinement, rather than providing a single answer, results in more accurate and thoughtful outputs. However, existing methods often rely heavily on…
Languages are shaped by the inductive biases of their users. Using a classical referential game, we investigate how artificial languages evolve when optimised for inductive biases in humans and large language models (LLMs) via Human-Human,…