2 papers
cs.AI2026
Convergent Emergence of In-Context Learning Across Modalities
Nathan Breslow, Seungwook Han, Daniel Hyunsoo Lee +3
Few-shot in-context learning (ICL), the capacity of a model to infer abstract patterns from input-output examples provided in its prompt and apply them to new inputs, has been exte…
cs.LG2026
SPICE: Simple Polysemantic Feature Interpretation via Clustering-based Explanation
Sehyun Lee, Dahee Kwon, Damin Lee +1
One of the pivotal recent challenges in neural network interpretability is polysemanticity, where a single neuron is activated by multiple, often unrelated concepts, hindering clea…