How to explain a DNN’s generalization ability and learning dynamics through the lens of interaction concepts? Our recent works ( https://t.co/aai8olRFAE, and https://t.co/5RUCekPMc5 in #NeurIPS2024) discover and theoretically prove a two-phase dynamics of interaction concepts

What is the mathematical mechanism shared by different attribution methods? Our TPAMI paper ( https://t.co/lGPF15Fua7) unifies 14 post-hoc attribution methods from the perspective of interactions. We prove that all 14 attribution methods can be represented as a weighted

Reasoning or memorization? Our recent work ( https://t.co/ouuAVs256Z) proposes a new axiomatic framework to formulate and quantify the exact reasoning and memorization effects in an LLM. See our project page for a brief introduction: https://t.co/YurNeLkrjw

Can the inference logic of a DNN be faithfully explained as symbolic concepts? Our #ICLR2024 paper ( https://t.co/6qn0euvUK7) makes an initial theoretical attempt to address this question. We prove that under three sufficient conditions, a DNN only encodes a small number of