Do Transformer Models Show Similar Attention Patterns to Task-Specific Human Gaze?

• URL: http://arxiv.org/abs/2205.10226v1
• Date: Mon, 25 Apr 2022 08:23:13 GMT
• Title: Do Transformer Models Show Similar Attention Patterns to Task-Specific Human Gaze?
• Authors: Stephanie Brandl, Oliver Eberle, Jonas Pilot, Anders S{\o}gaard
• Abstract summary: Self-attention functions in state-of-the-art NLP models often correlate with human attention. We investigate whether self-attention in large-scale pre-trained language models is as predictive of human eye fixation patterns during task-reading as classical cognitive models of human attention.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Learned self-attention functions in state-of-the-art NLP models often correlate with human attention. We investigate whether self-attention in large-scale pre-trained language models is as predictive of human eye fixation patterns during task-reading as classical cognitive models of human attention. We compare attention functions across two task-specific reading datasets for sentiment analysis and relation extraction. We find the predictiveness of large-scale pre-trained self-attention for human attention depends on `what is in the tail', e.g., the syntactic nature of rare contexts. Further, we observe that task-specific fine-tuning does not increase the correlation with human task-specific reading. Through an input reduction experiment we give complementary insights on the sparsity and fidelity trade-off, showing that lower-entropy attention vectors are more faithful.

Related papers

• Learning from Observer Gaze:Zero-Shot Attention Prediction Oriented by Human-Object Interaction Recognition [13.956664101032006]
  We first collect a novel gaze fixation dataset named IG, comprising 530,000 fixation points across 740 diverse interaction categories. We then introduce the zero-shot interaction-oriented attention prediction task ZeroIA, which challenges models to predict visual cues for interactions not encountered during training. Thirdly, we present the Interactive Attention model IA, designed to emulate human observers cognitive processes to tackle the ZeroIA problem.
  arXiv  Detail & Related papers  (2024-05-16T09:34:57Z)
• Closely Interactive Human Reconstruction with Proxemics and Physics-Guided Adaption [64.07607726562841]
  Existing multi-person human reconstruction approaches mainly focus on recovering accurate poses or avoiding penetration. In this work, we tackle the task of reconstructing closely interactive humans from a monocular video. We propose to leverage knowledge from proxemic behavior and physics to compensate the lack of visual information.
  arXiv  Detail & Related papers  (2024-04-17T11:55:45Z)
• Generating Human-Centric Visual Cues for Human-Object Interaction Detection via Large Vision-Language Models [59.611697856666304]
  Human-object interaction (HOI) detection aims at detecting human-object pairs and predicting their interactions. We propose three prompts with VLM to generate human-centric visual cues within an image from multiple perspectives of humans. We develop a transformer-based multimodal fusion module with multitower architecture to integrate visual cue features into the instance and interaction decoders.
  arXiv  Detail & Related papers  (2023-11-26T09:11:32Z)
• Eye vs. AI: Human Gaze and Model Attention in Video Memorability [22.718191366938278]
  We propose a Transformer-based model with naturalistic-temporal attention that matches SoTA performance on video memorability prediction. We compare model attention against human gaze fixation density maps collected through a small-scale eye-tracking experiment. We observe that the model assigns greater importance to the initial frames, mimicking temporal attention patterns found in humans.
  arXiv  Detail & Related papers  (2023-11-26T05:14:06Z)
• Attention cannot be an Explanation [99.37090317971312]
  We ask how effective are attention based explanations in increasing human trust and reliance in the underlying models? We perform extensive human study experiments that aim to qualitatively and quantitatively assess the degree to which attention based explanations are suitable. Our experiment results show that attention cannot be used as an explanation.
  arXiv  Detail & Related papers  (2022-01-26T21:34:05Z)
• Understanding top-down attention using task-oriented ablation design [0.22940141855172028]
  Top-down attention allows neural networks, both artificial and biological, to focus on the information most relevant for a given task. We aim to answer this with a computational experiment based on a general framework called task-oriented ablation design. We compare the performance of two neural networks, one with top-down attention and one without.
  arXiv  Detail & Related papers  (2021-06-08T21:01:47Z)
• Is Sparse Attention more Interpretable? [52.85910570651047]
  We investigate how sparsity affects our ability to use attention as an explainability tool. We find that only a weak relationship between inputs and co-indexed intermediate representations exists -- under sparse attention. We observe in this setting that inducing sparsity may make it less plausible that attention can be used as a tool for understanding model behavior.
  arXiv  Detail & Related papers  (2021-06-02T11:42:56Z)
• Gaze Perception in Humans and CNN-Based Model [66.89451296340809]
  We compare how a CNN (convolutional neural network) based model of gaze and humans infer the locus of attention in images of real-world scenes. We show that compared to the model, humans' estimates of the locus of attention are more influenced by the context of the scene.
  arXiv  Detail & Related papers  (2021-04-17T04:52:46Z)
• SparseBERT: Rethinking the Importance Analysis in Self-attention [107.68072039537311]
  Transformer-based models are popular for natural language processing (NLP) tasks due to its powerful capacity. Attention map visualization of a pre-trained model is one direct method for understanding self-attention mechanism. We propose a Differentiable Attention Mask (DAM) algorithm, which can be also applied in guidance of SparseBERT design.
  arXiv  Detail & Related papers  (2021-02-25T14:13:44Z)
• Attention Flows: Analyzing and Comparing Attention Mechanisms in Language Models [5.866941279460248]
  We propose a visual analytics approach to understanding fine-tuning in attention-based language models. Our visualization, Attention Flows, is designed to support users in querying, tracing, and comparing attention within layers, across layers, and amongst attention heads in Transformer-based language models.
  arXiv  Detail & Related papers  (2020-09-03T19:56:30Z)

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.