6 citations · 6 across the 10 of their papers we have counts for
10 papers
Surrogate Fidelity: When Can Open LLMs Explain Closed Ones?
Philippe Chlenski, Zachariah Carmichael, Ayush Warikoo +5
Mechanistic interpretability (MI) requires full access to model internals, yet the APIs for most widely deployed language models at best expose log-probabilities over output tokens…
Objective-Specific Privileged Bases via Full-Prefix Matryoshka Learning
Arghamitra Talukder, Philippe Chlenski, Itsik Pe'er
Learned representations are often invariant to rotational transformations, leaving individual dimensions non-identifiable and interchangeable. We study how Matryoshka Representatio…
Hyperbolic Genome Embeddings
Raiyan R. Khan, Philippe Chlenski, Itsik Pe'er
Current approaches to genomic sequence modeling often struggle to align the inductive biases of machine learning models with the evolutionarily-informed structure of biological sys…
Even Faster Hyperbolic Random Forests: A Beltrami-Klein Wrapper Approach
Philippe Chlenski, Itsik Pe'er
Decision trees and models that use them as primitives are workhorses of machine learning in Euclidean spaces. Recent work has further extended these models to the Lorentz model of…
Manify: A Python Library for Learning Non-Euclidean Representations
Philippe Chlenski, Kaizhu Du, Dylan Satow +2
We present Manify, an open-source Python library for non-Euclidean representation learning. Leveraging manifold learning techniques, Manify provides tools for learning embeddings i…
Mixed-curvature decision trees and random forests
Philippe Chlenski, Quentin Chu, Raiyan R. Khan +3
Decision trees (DTs) and their random forest (RF) extensions are workhorses of classification and regression in Euclidean spaces. However, algorithms for learning in non-Euclidean…
Transcoders Find Interpretable LLM Feature Circuits
Jacob Dunefsky, Philippe Chlenski, Neel Nanda
A key goal in mechanistic interpretability is circuit analysis: finding sparse subgraphs of models corresponding to specific behaviors or capabilities. However, MLP sublayers make…
Mixed-Curvature Decision Trees and Random Forests
Philippe Chlenski, Quentin Chu, Itsik Pe'er
We extend decision tree and random forest algorithms to product space manifolds: Cartesian products of Euclidean, hyperspherical, and hyperbolic manifolds. Such spaces have extreme…
Variational Pseudo Marginal Methods for Jet Reconstruction in Particle Physics
Hanming Yang, Antonio Khalil Moretti, Sebastian Macaluso +3
Reconstructing jets, which provide vital insights into the properties and histories of subatomic particles produced in high-energy collisions, is a main problem in data analyses in…
Fast hyperboloid decision tree algorithms
Philippe Chlenski, Ethan Turok, Antonio Moretti +1
Hyperbolic geometry is gaining traction in machine learning for its effectiveness at capturing hierarchical structures in real-world data. Hyperbolic spaces, where neighborhoods gr…