conftrace_

flip rate

2 papers

Explore in graph

Co-occurring keywords · papers mentioning both

policy parameter dimension (1) masked decay term (1) dense fine-tuning (1) dataset size (1) 2:4 sparsity (1) feed-forward network (1) noise robustness (1) bradley-terry-luce model (1) sparsity (1) direct preference optimization (1)

Papers

Accelerating Transformer Pre-training with 2:4 Sparsity ICML 2024 Provably Robust DPO: Aligning Language Models with Noisy Feedback ICML 2024