Bayesian methods have undeniable value, e.g. in data analysis, but they play a peripheral role in “Modern AI” (i.e. deep learning), which prioritises scalable, empirical approaches over theoretical grounding---a trend that is unlikely to change.
As language models become increasingly sophisticated and existing benchmarks approach saturation, the need for rigorous evaluation methods grows more pressing. Many evaluations lack the statistical rigour needed to draw meaningful conclusions, leading to a potential over-confidence in results that might not hold up under scrutiny or replication. This post advocates for bringing fundamental statistical principles to language model evaluation, demonstrating how basic statistical analysis can provide more reliable insights into model capabilities and limitations.
There is huge value in developing new benchmarks and I think the one that proposed in the paper 'GSM-Symbolic: Understanding the Limitations of Mathematical Reasoning in Large Language Models' by Apple is quite neat and useful! The accompanying analysis, in my opinion, can be substantially improved with the help of basic statistics. Without those we risk over-interpreting results and drawing misleading conclusions. I never thought I would be the one advocating for the use of hypothesis tests and p-values, but here we are... When it comes to language models evals, it is time to make statistics great again!
A simple tutorial on how to implement static BED in BayesFlow
Machine learning, and deep probabilistic modelling specifically, seems to be revolutionising the space of data compression. This short post describes 1) the basic components of the data compression pipeline; 2) the objective used to optimise model parameters and its equivalence to training a VAE; and 3) some of the challenges that need to be solved.
Bayesian Optimal Experimental Design (BOED) is an elegant mathematical framework that enables us to design experiments optimally. This introductory post describes the BOED framework and the computational challenges associated with deploying it in applications.