SPIN: Sparsifying and Integrating Internal Neurons in Large Language Models for Text Classification

• URL: http://arxiv.org/abs/2311.15983v2
• Date: Wed, 5 Jun 2024 17:15:47 GMT
• Title: SPIN: Sparsifying and Integrating Internal Neurons in Large Language Models for Text Classification
• Authors: Difan Jiao, Yilun Liu, Zhenwei Tang, Daniel Matter, Jürgen Pfeffer, Ashton Anderson,
• Abstract summary: We present a model-agnostic framework that sparsifies and integrates internal neurons of intermediate layers of Large Language Models for text classification. SPIN significantly improves text classification accuracy, efficiency, and interpretability.
• Score: 6.227343685358882
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Among the many tasks that Large Language Models (LLMs) have revolutionized is text classification. Current text classification paradigms, however, rely solely on the output of the final layer in the LLM, with the rich information contained in internal neurons largely untapped. In this study, we present SPIN: a model-agnostic framework that sparsifies and integrates internal neurons of intermediate layers of LLMs for text classification. Specifically, SPIN sparsifies internal neurons by linear probing-based salient neuron selection layer by layer, avoiding noise from unrelated neurons and ensuring efficiency. The cross-layer salient neurons are then integrated to serve as multi-layered features for the classification head. Extensive experimental results show our proposed SPIN significantly improves text classification accuracy, efficiency, and interpretability.

Related papers

• Improving Neuron-level Interpretability with White-box Language Models [11.898535906016907]
  We introduce a white-box transformer-like architecture named Coding RAte TransformEr (CRATE) Our comprehensive experiments showcase significant improvements (up to 103% relative improvement) in neuron-level interpretability. CRATE's increased interpretability comes from its enhanced ability to consistently and distinctively activate on relevant tokens.
  arXiv  Detail & Related papers  (2024-10-21T19:12:33Z)
• Unveiling the Power of Sparse Neural Networks for Feature Selection [60.50319755984697]
  Sparse Neural Networks (SNNs) have emerged as powerful tools for efficient feature selection. We show that SNNs trained with dynamic sparse training (DST) algorithms can achieve, on average, more than $50%$ memory and $55%$ FLOPs reduction. Our findings show that feature selection with SNNs trained with DST algorithms can achieve, on average, more than $50%$ memory and $55%$ FLOPs reduction.
  arXiv  Detail & Related papers  (2024-08-08T16:48:33Z)
• Sharing Matters: Analysing Neurons Across Languages and Tasks in LLMs [70.3132264719438]
  We aim to fill the research gap by examining how neuron activation is shared across tasks and languages. We classify neurons into four distinct categories based on their responses to a specific input across different languages. Our analysis reveals the following insights: (i) the patterns of neuron sharing are significantly affected by the characteristics of tasks and examples; (ii) neuron sharing does not fully correspond with language similarity; (iii) shared neurons play a vital role in generating responses, especially those shared across all languages.
  arXiv  Detail & Related papers  (2024-06-13T16:04:11Z)
• A Hybrid Neural Coding Approach for Pattern Recognition with Spiking Neural Networks [53.31941519245432]
  Brain-inspired spiking neural networks (SNNs) have demonstrated promising capabilities in solving pattern recognition tasks. These SNNs are grounded on homogeneous neurons that utilize a uniform neural coding for information representation. In this study, we argue that SNN architectures should be holistically designed to incorporate heterogeneous coding schemes.
  arXiv  Detail & Related papers  (2023-05-26T02:52:12Z)
• Finding Neurons in a Haystack: Case Studies with Sparse Probing [2.278231643598956]
  Internal computations of large language models (LLMs) remain opaque and poorly understood. We train $k$-sparse linear classifiers to predict the presence of features in the input. By varying the value of $k$ we study the sparsity of learned representations and how this varies with model scale.
  arXiv  Detail & Related papers  (2023-05-02T17:13:55Z)
• Neuro-symbolic Rule Learning in Real-world Classification Tasks [75.0907310059298]
  We extend pix2rule's neural DNF module to support rule learning in real-world multi-class and multi-label classification tasks. We propose a novel extended model called neural DNF-EO (Exactly One) which enforces mutual exclusivity in multi-class classification.
  arXiv  Detail & Related papers  (2023-03-29T13:27:14Z)
• Supervised Feature Selection with Neuron Evolution in Sparse Neural Networks [17.12834153477201]
  We propose a novel resource-efficient supervised feature selection method using sparse neural networks. By gradually pruning the uninformative features from the input layer of a sparse neural network trained from scratch, NeuroFS derives an informative subset of features efficiently. NeuroFS achieves the highest ranking-based score among the considered state-of-the-art supervised feature selection models.
  arXiv  Detail & Related papers  (2023-03-10T17:09:55Z)
• Discovering Salient Neurons in Deep NLP Models [31.18937787704794]
  We present a technique called as Linguistic Correlation Analysis to extract salient neurons in the model. Our data-driven, quantitative analysis illuminates interesting findings. Our code is publicly available as part of the NeuroX toolkit.
  arXiv  Detail & Related papers  (2022-06-27T13:31:49Z)
• Natural Language Descriptions of Deep Visual Features [50.270035018478666]
  We introduce a procedure that automatically labels neurons with open-ended, compositional, natural language descriptions. We use MILAN for analysis, characterizing the distribution and importance of neurons selective for attribute, category, and relational information in vision models. We also use MILAN for auditing, surfacing neurons sensitive to protected categories like race and gender in models trained on datasets intended to obscure these features.
  arXiv  Detail & Related papers  (2022-01-26T18:48:02Z)
• Analyzing Individual Neurons in Pre-trained Language Models [41.07850306314594]
  We find small subsets of neurons to predict linguistic tasks, with lower level tasks localized in fewer neurons, compared to higher level task of predicting syntax. For example, we found neurons in XLNet to be more localized and disjoint when predicting properties compared to BERT and others, where they are more distributed and coupled.
  arXiv  Detail & Related papers  (2020-10-06T13:17:38Z)
• Non-linear Neurons with Human-like Apical Dendrite Activations [81.18416067005538]
  We show that a standard neuron followed by our novel apical dendrite activation (ADA) can learn the XOR logical function with 100% accuracy. We conduct experiments on six benchmark data sets from computer vision, signal processing and natural language processing.
  arXiv  Detail & Related papers  (2020-02-02T21:09:39Z)

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.