Skip to content

Interpretable weight-space programming builds transparent Transformers and brain encoding models

Interpretable weight-space programming builds transparent transformers and brain encoding models

arXiv· arXiv:2611.18114· Yash Mehta, Chandan Singh, Weijia Xu, Richard J. Antonello, Adeen Flinker, Nima Mesgarani, Michael F. Bonner, Nebojsa Jojic, Jianfeng Gao·Indexed by this site · 17 hr ago· Originally publishedAlignment
AI summary

The authors propose interpretable weight-space programming (iWSP), in which LLM agents iteratively build interpretable models by writing model weights and features directly in code without backpropagation; the agents must rely on prior knowledge and reason about past trials, proposing hypotheses, manually constructing and improving human-interpretable computations, and an LLM audit confirmed the interpretability of these hand-built models. The authors first had iWSP build compact and accurate Transformers on benchmarks such as arithmetic and sentiment classification, then applied it to neural encoding models of vision, language and hearing: on visual cortex it outperformed existing interpretable baselines in prediction, revealed different computations in early and higher visual areas, and generalized to unseen data; in language fMRI it revealed confounders in the data, and in auditory ECoG it systematically found interpretable acoustic features that predict neural activity. The authors say these results establish iWSP as a general paradigm combining high predictive performance with human-interpretable scientific discovery.

Related events and news

1 events · 1 news mentions

Read originalarxiv.org