Linking forward-pass dynamics in Transformers and real-time human processing

• URL: http://arxiv.org/abs/2504.14107v1
• Date: Fri, 18 Apr 2025 23:38:14 GMT
• Title: Linking forward-pass dynamics in Transformers and real-time human processing
• Authors: Jennifer Hu, Michael A. Lepori, Michael Franke,
• Abstract summary: We investigate the link between real-time processing in humans and "layer-time" dynamics in Transformer models.<n>Our results suggest that Transformer processing and human processing may be facilitated or impeded by similar properties of an input stimulus.
• Score: 6.165163123577484
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Modern AI models are increasingly being used as theoretical tools to study human cognition. One dominant approach is to evaluate whether human-derived measures (such as offline judgments or real-time processing) are predicted by a model's output: that is, the end-product of forward pass(es) through the network. At the same time, recent advances in mechanistic interpretability have begun to reveal the internal processes that give rise to model outputs, raising the question of whether models and humans might arrive at outputs using similar "processing strategies". Here, we investigate the link between real-time processing in humans and "layer-time" dynamics in Transformer models. Across five studies spanning domains and modalities, we test whether the dynamics of computation in a single forward pass of pre-trained Transformers predict signatures of processing in humans, above and beyond properties of the model's output probability distribution. We consistently find that layer-time dynamics provide additional predictive power on top of output measures. Our results suggest that Transformer processing and human processing may be facilitated or impeded by similar properties of an input stimulus, and this similarity has emerged through general-purpose objectives such as next-token prediction or image recognition. Our work suggests a new way of using AI models to study human cognition: not just as a black box mapping stimuli to responses, but potentially also as explicit processing models.

Related papers

• Transcendence: Generative Models Can Outperform The Experts That Train Them [55.885802048647655]
  We study the phenomenon of transcendence: when a generative model achieves capabilities that surpass the abilities of the experts generating its data. We demonstrate transcendence by training an autoregressive transformer to play chess from game transcripts, and show that the trained model can sometimes achieve better performance than all players in the dataset.
  arXiv  Detail & Related papers  (2024-06-17T17:00:52Z)
• Humanoid Locomotion as Next Token Prediction [84.21335675130021]
  Our model is a causal transformer trained via autoregressive prediction of sensorimotor trajectories. We show that our model enables a full-sized humanoid to walk in San Francisco zero-shot. Our model can transfer to the real world even when trained on only 27 hours of walking data, and can generalize commands not seen during training like walking backward.
  arXiv  Detail & Related papers  (2024-02-29T18:57:37Z)
• TransFusion: A Practical and Effective Transformer-based Diffusion Model for 3D Human Motion Prediction [1.8923948104852863]
  We propose TransFusion, an innovative and practical diffusion-based model for 3D human motion prediction. Our model leverages Transformer as the backbone with long skip connections between shallow and deep layers. In contrast to prior diffusion-based models that utilize extra modules like cross-attention and adaptive layer normalization, we treat all inputs, including conditions, as tokens to create a more lightweight model.
  arXiv  Detail & Related papers  (2023-07-30T01:52:07Z)
• Learning Theory of Mind via Dynamic Traits Attribution [59.9781556714202]
  We propose a new neural ToM architecture that learns to generate a latent trait vector of an actor from the past trajectories. This trait vector then multiplicatively modulates the prediction mechanism via a fast weights' scheme in the prediction neural network. We empirically show that the fast weights provide a good inductive bias to model the character traits of agents and hence improves mindreading ability.
  arXiv  Detail & Related papers  (2022-04-17T11:21:18Z)
• HiT-DVAE: Human Motion Generation via Hierarchical Transformer Dynamical VAE [37.23381308240617]
  We propose Hierarchical Transformer Dynamical Variational Autoencoder, HiT-DVAE, which implements auto-regressive generation with transformer-like attention mechanisms. We evaluate the proposed method on HumanEva-I and Human3.6M with various evaluation methods, and outperform the state-of-the-art methods on most of the metrics.
  arXiv  Detail & Related papers  (2022-04-04T15:12:34Z)
• STAR: Sparse Transformer-based Action Recognition [61.490243467748314]
  This work proposes a novel skeleton-based human action recognition model with sparse attention on the spatial dimension and segmented linear attention on the temporal dimension of data. Experiments show that our model can achieve comparable performance while utilizing much less trainable parameters and achieve high speed in training and inference.
  arXiv  Detail & Related papers  (2021-07-15T02:53:11Z)
• Probabilistic Human Motion Prediction via A Bayesian Neural Network [71.16277790708529]
  We propose a probabilistic model for human motion prediction in this paper. Our model could generate several future motions when given an observed motion sequence. We extensively validate our approach on a large scale benchmark dataset Human3.6m.
  arXiv  Detail & Related papers  (2021-07-14T09:05:33Z)
• Multimodal Deep Generative Models for Trajectory Prediction: A Conditional Variational Autoencoder Approach [34.70843462687529]
  We provide a self-contained tutorial on a conditional variational autoencoder approach to human behavior prediction. The goals of this tutorial paper are to review and build a taxonomy of state-of-the-art methods in human behavior prediction.
  arXiv  Detail & Related papers  (2020-08-10T03:18:27Z)
• Learning Predictive Models From Observation and Interaction [137.77887825854768]
  Learning predictive models from interaction with the world allows an agent, such as a robot, to learn about how the world works. However, learning a model that captures the dynamics of complex skills represents a major challenge. We propose a method to augment the training set with observational data of other agents, such as humans.
  arXiv  Detail & Related papers  (2019-12-30T01:10:41Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.