1 citations · 1 across the 7 of their papers we have counts for
7 papers
How Do Language Models Choose Between Context and Memory?
Benjamin Shih, John Winnicki, Arianna Cao
When contextual information conflicts with knowledge stored in model parameters, activation directions can be used to decode and steer which source the model follows. However, succ…
When Does Activation Steering Change What a Model Computes From?
Benjamin Shih, John Winnicki, Eric Darve
Activation steering can reliably change an agent's output by modifying its internal activations. Yet arriving at the same answer need not involve the same computation: behavioral e…
Domain-Filtered Knowledge Graphs from Sparse Autoencoder Features
John Winnicki, Abeynaya Gnanasekaran, Eric Darve
Sparse autoencoders (SAEs) extract millions of interpretable features from a language model, but flat feature inventories aren't very useful on their own. Domain concepts get mixed…
Flash-SD-KDE: Accelerating SD-KDE with Tensor Cores
Elliot L. Epstein, Rajat Vadiraj Dwaraknath, John Winnicki
Score-debiased kernel density estimation (SD-KDE) achieves improved asymptotic convergence rates over classical KDE, but its use of an empirical score has made it significantly slo…
Allocate Marginal Reviews to Borderline Papers Using LLM Comparative Ranking
Elliot L. Epstein, Rajat Dwaraknath, John Winnicki +1
This paper argues that large ML conferences should allocate marginal review capacity primarily to papers near the acceptance boundary, rather than spreading extra reviews via rando…
LLMs are Overconfident: Evaluating Confidence Interval Calibration with FermiEval
Elliot L. Epstein, John Winnicki, Thanawat Sornwanee +1
Large language models (LLMs) excel at numerical estimation but struggle to correctly quantify uncertainty. We study how well LLMs construct confidence intervals around their own an…
SD-KDE: Score-Debiased Kernel Density Estimation
Elliot L. Epstein, Rajat Dwaraknath, Thanawat Sornwanee +2
We propose a novel method for density estimation that leverages an estimated score function to debias kernel density estimation (SD-KDE). In our approach, each data point is adjust…