Interpretability

On Behalf of the Stakeholders: Trends in NLP Model Interpretability in the Era of LLMs NAACL 2025 One fish, two fish, but not the whole sea: Alignment reduces language models’ conceptual diversity NAACL 2025 One Hundred Neural Networks and Brains Watching Videos: Lessons from Alignment ICLR 2025 On Explaining Equivariant Graph Networks via Improved Relevance Propagation ICML 2025 On Explaining Knowledge Distillation: Measuring and Visualising the Knowledge Transfer Process WACV 2025 On Localizing and Deleting Toxic Memories in Large Language Models NAACL 2025 On-Policy Self-Alignment with Fine-grained Knowledge Feedback for Hallucination Mitigation ACL 2025 On Reducing Factual Hallucinations in Graph-to-Text Generation Using Large Language Models COLING 2025 On Relation-Specific Neurons in Large Language Models EMNLP 2025 On Support Samples of Next Word Prediction ACL 2025 On the Complexity-Faithfulness Trade-off of Gradient-Based Explanations ICCV 2025 On the Consistency of Commonsense in Large Language Models ACL 2025 On the Consistency of Video Large Language Models in Temporal Comprehension CVPR 2025 On the Convergence of Moral Self-Correction in Large Language Models IJCNLP 2025 On the Feasibility of In-Context Probing for Data Attribution NAACL 2025 On the Generalization of Representation Uncertainty in Earth Observation ICCV 2025 On the Hölder Stability of Multiset and Graph Neural Networks ICLR 2025 On the Mutual Influence of Gender and Occupation in LLM Representations ACL 2025 On the reliability of feature attribution methods for speech classification Interspeech 2025 On the Representations of Entities in Auto-regressive Large Language Models EMNLP 2025 On the Role of Attention Heads in Large Language Model Safety ICLR 2025 On the Versatility of Sparse Autoencoders for In-Context Learning EMNLP 2025 Open Domain Question Answering with Conflicting Contexts NAACL 2025 OpenRCA: Can Large Language Models Locate the Root Cause of Software Failures? ICLR 2025 OpenViewer: Openness-Aware Multi-View Learning AAAI 2025