AI·Procrustes-conditioned Joint End-to-end Top-K Sparse Autoencoders extract cross-seed universal features from independently trained BERT models to address dictionary learning misalignment in mechanistic interpretability.
Search
What are you looking for?
4 results for “interpretability”
AI·User questions whether to withdraw from ACL ARR and resubmit to a workshop after mediocre 2.5-3 scores on an Interpretability-track EMNLP paper.
Multimodal large language models (MLLMs) generate responses autoregressively, integrating visual and linguistic information in an evolving context. Prior work on interpretability has focused on individual layers and circuits (where), leaving the token-level dynamics of multimodal computation during generation (when) underexplored. We address this gap and study attention shifts as per semantic role; tracking model attention to image, text, instruction, and previously generated tokens, One Token …
A key challenge in Arabic NLP is the scarcity of dialectal data relative to Modern Standard Arabic (MSA), causing LLMs to overproduce MSA and struggle with dialectally accurate generation. From an interpretability perspective, this raises a fundamental question: where and how are dialectal features encoded within model internals, and can these representations be leveraged to improve dialect generation without fine-tuning? This study investigates two complementary inference-time approaches that …