| Challenge: | Existing studies on controlling neuron distribution for interpretability have focused on focusing on monosemanticity instead of focusing solely on feature interactions. |
| Approach: | They propose a method to regularize feature superposition by encoding representations of multiple features within a single neuron. |
| Outcome: | The proposed method improves model interpretability without compromising prediction performance. |
Similar Papers
Constructing Interpretable Features from Compositional Neuron Groups (2026.acl-long)
Copied to clipboard
| Challenge: | Existing methods for analyzing LLMs rely on dictionary learning with sparse autoencoders (SAEs) however, SAEs struggle in causal evaluations and lack intrinsic interpretability, as their learning is not explicitly tied to the computations of the model. |
| Approach: | They propose to decompose MLP activations with semi-nonnegative matrix factorization (SNMF) such that the learned features are mapped to their activating inputs, making them directly interpretable. |
| Outcome: | Experiments on Llama 3.1, Gemma 2 and GPT-2 show that SNMF derived features outperform SAEs and a strong supervised baseline on causal steering while aligning with human-interpretable concepts. |
DoLFIn: Distributions over Latent Features for Interpretability (2020.coling-main)
Copied to clipboard
| Challenge: | Existing approaches to interpret neural networks face a trade-off between a model's usefulness and its complexity. |
| Approach: | They propose a novel approach to achieve interpretability that avoids this trade-off by using probability as the central quantity instead of a fixed quantity. |
| Outcome: | The proposed approach outperforms the classical CNN and BiLSTM classifiers on the SST2 and AG-news datasets. |
Improve Interpretability of Neural Networks via Sparse Contrastive Coding (2022.findings-emnlp)
Copied to clipboard
| Challenge: | XAI has achieved remarkable advances, but few efforts have been devoted to solving the problem. |
| Approach: | They propose a model-agnostic explanation method termed Sparse Contrastive Coding . they use model-based explanations to explain the black-box in a more model-oriented way . |
| Outcome: | The proposed method outperforms five state-of-the-art methods in interpretability and classification metrics. |
Multi-layer Representation Fusion for Neural Machine Translation (C18-1)
Copied to clipboard
| Challenge: | Neural machine translation systems require a number of stacked layers for deep models, but the prediction depends on the sentence representation of the top-most layer with no access to low-level representations. |
| Approach: | They propose a multi-layer representation fusion approach to fusing stacked layers to learn a better representation from the stack. |
| Outcome: | The proposed approach yields 0.92 and 0.56 BLEU points over the strong Transformer baseline on IWSLT German-English and NIST Chinese-English MT tasks respectively. |
Evaluating Sparse Autoencoders for Monosemantic Representation (2026.findings-eacl)
Copied to clipboard
| Challenge: | Sparse autoencoders (SAEs) have been proposed to mitigate polysemanticity, where neurons activate for multiple unrelated concepts. |
| Approach: | They propose a sparse autoencoder to transform dense activations into sparser, more interpretable features by transforming them into sparses. |
| Outcome: | The proposed model reduces polysemanticity and achieves higher concept separability. |
Sparsifying Transformer Models with Trainable Representation Pooling (2022.acl-long)
Copied to clipboard
| Challenge: | Existing approaches to sparsify attention in the Transformer model are based on quadratic memory complexity and a lack of information for each word. |
| Approach: | They propose a method to sparsify attention in a Transformer model by learning to select the most-informative token representations during the training process. |
| Outcome: | The proposed model performs better than the current SOTA model while being 1.8 faster during training, 4.5 faster inference and 13 more efficient in the decoder. |
AligNART: Non-autoregressive Neural Machine Translation by Jointly Learning to Estimate Alignment and Translate (2021.emnlp-main)
Copied to clipboard
| Challenge: | Non-autoregressive neural machine translation models suffer from the multi-modality problem . aligNART leverages full alignment information to explicitly reduce the modality of the target distribution . |
| Approach: | They propose an alignment decomposition method which explicitly reduces the modality of the target distribution. |
| Outcome: | The proposed model outperforms previous models that focus on modality reduction on two translation tasks. |
Soft Syntactic Reinforcement for Neural Event Extraction (2025.naacl-long)
Copied to clipboard
| Challenge: | Recent event extraction methods rely on pre-trained language models but still suffer from errors due to a lack of syntactic knowledge. |
| Approach: | They propose a method to incorporate syntactic information into PLM-based models for event extraction (EE) this method uses a standard dependency corpus to select syntax-related dimensions of the model's representation. |
| Outcome: | The proposed method outperforms baseline models and existing syntactic reinforcement methods on sentence-level and document-level EE benchmark datasets. |
Transformer-specific Interpretability (2024.eacl-tutorials)
Copied to clipboard
| Challenge: | Transformers are dominant play-ers in various scientific fields, but their inner workings remain opaque. |
| Approach: | This tutorial presents a trending approach to interpreting Transformers . it uses specific features of the Transformer architecture to quantify context- mixing interactions . |
| Outcome: | This tutorial aims to show how a new trending approach can be applied to Transformer-based models. |
Learning to Recover from Multi-Modality Errors for Non-Autoregressive Neural Machine Translation (2020.acl-main)
Copied to clipboard
| Challenge: | Existing non-autoregressive neural machine translation models suffer from multi-modality problem . despite their autoregressivity, most NMT models suffer with slow decoding speed . |
| Approach: | They propose a semi-autoregressive model which generates a translation as a sequence of segments while each segment is predicted token-by-token. |
| Outcome: | The proposed model can achieve 4 times speedup while maintaining comparable performance. |