Papers by Miles Gilberti

2 papers
Sparse Feature Coactivation Reveals Causal Semantic Modules in Large Language Models (2026.acl-long)

Copied to clipboard

Challenge: Recent work has focused on layerwise interpretations, lacking fine-grained interpretation of specific features and their interaction.
Approach: They identify semantically coherent, context-consistent network components in large language models . they use sparse autoencoders to coactivate sparsity features from a handful of prompts .
Outcome: The proposed model can capture concepts and relations more comprehensively than individual features while maintaining specificity.
Discovering Properties of Inflectional Morphology in Neural Emergent Communication (2026.acl-long)

Copied to clipboard

Challenge: Emergent communication studies protocols developed between two or more deep neural network-based agents . common evaluation metrics for large-vocabulary setting are overly simplified .
Approach: They propose to reinterpret an EmCom setting by imposing a small-vocabulary constraint to simulate double articulation and formulating a novel setting analogous to naturalistic inflectional morphology.
Outcome: The proposed model favors protocols that represent attributes with unique characters and compose them syntactically.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations