LinguaLens: Towards Interpreting Linguistic Mechanisms of Large Language Models via Sparse Auto-Encoder (2025.emnlp-main)
Copied to clipboard
| Challenge: | Prior research on linguistic mechanisms of large language models is limited by coarse granularity, limited analysis scale, and narrow focus. |
| Approach: | They propose a framework for analyzing the linguistic mechanisms of large language models based on Sparse Auto-Encoders. |
| Outcome: | The proposed framework extracts Chinese and English linguistic features across four dimensions . it uncovers intrinsic representations of linguistic knowledge in LLMs and can control outputs . |
Similar Papers
A Survey on Sparse Autoencoders: Interpreting the Internal Mechanisms of Large Language Models (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Sparse Autoencoders (SAEs) can disentangle complex features into more interpretable components. |
| Approach: | They propose to use Sparse Autoencoders to disentangle LLM features into more interpretable components. |
| Outcome: | The proposed method disentangles complex features into more interpretable components. |
Unveiling Language-Specific Features in Large Language Models via Sparse Autoencoders (2025.acl-long)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) exhibit impressive abilities in various domains such as text generation, instruction following, and reasoning. |
| Approach: | They propose a method to decompose the activations of Large Language Models into a sparse linear combination of SAE features. |
| Outcome: | The proposed method shows that some features are strongly related to specific languages, while others are unaffected by ablating them. |
Deciphering Cultural Representations in Large Language Models via Sparse Autoencoders (2026.findings-acl)
Copied to clipboard
| Challenge: | Prior work has identified so-called cultural neurons, but individual neurons are often polysemous, conflating abstract cultural knowledge with surface-level lexical cues due to superposition. |
| Approach: | They apply Sparse Autoencoders to decompose LLM activations into sparse, interpretable feature representations that disentangle culturally selective features. |
| Outcome: | The proposed model disentangles culturally selective features from paraphrasing and task formats, indicating abstraction beyond lexical correlations. |
Sparse Feature Coactivation Reveals Causal Semantic Modules in Large Language Models (2026.acl-long)
Copied to clipboard
Ruixuan Deng, Xiaoyang Hu, Miles Gilberti, Shane Storks, Aman Taxali, Mike Angstadt, Chandra Sripada, Joyce Chai
| Challenge: | Recent work has focused on layerwise interpretations, lacking fine-grained interpretation of specific features and their interaction. |
| Approach: | They identify semantically coherent, context-consistent network components in large language models . they use sparse autoencoders to coactivate sparsity features from a handful of prompts . |
| Outcome: | The proposed model can capture concepts and relations more comprehensively than individual features while maintaining specificity. |
Data and Model Centric Approaches for Expansion of Large Language Models to New languages (2025.emnlp-tutorials)
Copied to clipboard
| Challenge: | Existing LLMs mainly support English alongside a handful of high resource languages . this leaves a major gap for most low-resource languages despite increasing pace of research . |
| Approach: | This tutorial examines approaches to expand the language coverage of LLMs . they look at tokenizer training, pre-training, instruction tuning, alignment, evaluation, etc. |
| Outcome: | This tutorial examines approaches to expand the language coverage of LLMs . it provides an efficient and viable path to bring LLM technologies to low-resource languages . |
Large Language Models Share Representations of Latent Grammatical Concepts Across Typologically Diverse Languages (2025.naacl-long)
Copied to clipboard
| Challenge: | In the brains of human bilinguals, syntax processing may occur in similar regions for their first and second language, depending on factors like when the second language was learned and language proficiency. |
| Approach: | They propose to use sparse autoencoders to train Llama-3-8B and Aya-23-8B models to train multilingual models that share morphsyntactic representations of grammatical concepts. |
| Outcome: | The proposed model can predict plural verbs in different languages by activating the same plural feature. |
Extracting Linguistic Information from Large Language Models: Syntactic Relations and Derivational Knowledge (2025.emnlp-main)
Copied to clipboard
| Challenge: | Using large language models, we study their morphosyntactic competence and generalization capabilities. |
| Approach: | They propose to use morphosyntactic tasks to study their linguistic knowledge and generalization capabilities to extract different types of morphological structure for typologically diverse languages. |
| Outcome: | The proposed models outperform GPT-4o and LLaMA 3.3-70B in all diagnostic tasks, but show little evidence of abstract morphological rule learning. |
SAGE: An Agentic Explainer Framework for Interpreting SAE Features in Language Models (2026.eacl-industry)
Copied to clipboard
| Challenge: | Large language models (LLMs) have achieved remarkable progress, yet their internal mechanisms remain largely opaque. |
| Approach: | They propose an agent-based framework that recasts feature interpretation from a passive, single-pass generation task into an explanation-driven process. |
| Outcome: | The proposed framework produces explanations with significantly higher generative and predictive accuracy compared to state-of-the-art baselines. |
Route Sparse Autoencoder to Interpret Large Language Models (2025.emnlp-main)
Copied to clipboard
| Challenge: | Sparse autoencoders (SAEs) extract interpretable and monosemantic features in large language models . prior work focused on feature extraction from a single layer, failing to capture activations that span multiple layers. |
| Approach: | They propose a framework that integrates a routing mechanism with a shared SAE to efficiently extract features from multiple layers. |
| Outcome: | The proposed framework extracts features from multiple layers while incurring minimal parameter overhead while achieving high interpretability and flexibility. |
The Linguistic Connectivities Within Large Language Models (2025.findings-acl)
Copied to clipboard
Dan Wang, Boxi Cao, Ning Bian, Xuanang Chen, Yaojie Lu, Hongyu Lin, Jia Zheng, Le Sun, Shanshan Jiang, Bin Dong, Xianpei Han
| Challenge: | Recent studies have discovered notable disparities in their performance across different languages. |
| Approach: | They conduct a systematic investigation into the behaviors of large language models across 27 different languages on 3 different scenarios and reveals a Linguistic Map correlates with the richness of available resources and linguistic family relations. |
| Outcome: | The proposed model demonstrates that there are significant disparities in performance across languages across 27 different languages on 3 different scenarios. |