4 papers
Slaying the Hydra: Interaction-Aware Circuit Discovery in Language Models
Sankaran Vaidyanathan, Rafal Urbaniak, Emily Bunnapradist +2
Localizing behavior to individual components of a language model is a central goal of mechanistic interpretability. However, scoring components one at a time misses context-depende…
A Computationally Feasible Framework for Causal Probabilistic Explanation
Rafal Urbaniak, Sam Witty, Daniel Waxman +7
Explaining why a specific outcome occurred, and which inputs deserve the blame or credit, is central to philosophical, scientific, and policy analysis. Existing tools split into tw…
A Bayesian approach to uncertainty in word embedding bias estimation
Alicja Dobrzeniecka, Rafal Urbaniak
Multiple measures, such as WEAT or MAC, attempt to quantify the magnitude of bias present in word embeddings in terms of a single-number metric. However, such metrics and the relat…
Reconciling Bayesian Epistemology and Narration-based Approaches to Judiciary Fact-finding
Rafal Urbaniak
Legal probabilism (LP) claims the degrees of conviction in juridical fact-finding are to be modeled exactly the way degrees of beliefs are modeled in standard bayesian epistemology…