Next Token Perception Score: Analytical Assessment of your LLM Perception Skills

TL;DR

This paper introduces the Next Token Perception Score (NTPS), a metric to evaluate how well autoregressive language model representations align with perception tasks, correlating strongly with downstream performance and aiding in fine-tuning assessments.

Contribution

The paper proposes NTPS, a novel analytical metric for measuring the alignment between autoregressive representations and perception tasks, validated across multiple models and datasets.

Findings

NTPS correlates strongly with linear probe accuracy.

LoRA fine-tuning increases NTPS, improving perception alignment.

NTPS predicts gains from LoRA fine-tuning.

Abstract

Autoregressive pretraining has become the de facto paradigm for learning general-purpose representations in large language models (LLMs). However, linear probe performance across downstream perception tasks shows substantial variability, suggesting that features optimized for next-token prediction do not consistently transfer well to downstream perception tasks. We demonstrate that representations learned via autoregression capture features that may lie outside the subspaces most informative for perception. To quantify the (mis)alignment between autoregressive pretraining and downstream perception, we introduce the Next Token Perception Score (NTPS)-a score derived under a linear setting that measures the overlap between autoregressive and perception feature subspaces. This metric can be easily computed in closed form from pretrained representations and labeled data, and is proven to both upper- and lower-bound the excess loss. Empirically, we show that NTPS correlates strongly with linear probe accuracy across 12 diverse NLP datasets and eight pretrained models ranging from 270M to 8B parameters, confirming its utility as a measure of alignment. Furthermore, we show that NTPS increases following low-rank adaptation (LoRA) fine-tuning, especially in large models, suggesting that LoRA aligning representations to perception tasks enhances subspace overlap and thus improves downstream performance. More importantly, we find that NTPS reliably predicts the additional accuracy gains attained by LoRA finetuning thereby providing a lightweight prescreening tool for LoRA adaptation. Our results offer both theoretical insights and practical tools for analytically assessing LLM perception skills.