🩵 123

Alex

snaykey

3 7

·

AI & ML interests

AI Safety & Redteaming

Recent Activity

updated a Space about 2 hours ago

snaykey/repro-swift-svd-theoretical-optimality-meets-practical-efficiency-in-low-rank-llm-compression

updated a Space about 2 hours ago

snaykey/repro-optimal-regret-policy-optimization-contextual-bandits

updated a Space about 3 hours ago

snaykey/repro-projected-ssgd

View all activity

Organizations

snaykey 's Spaces 172

Repro: Swift-SVD — Low-Rank LLM Compression

Explore and share Swift‑SVD experiment logs

Reproduction: Optimal Regret for Policy Optimization in Contextual Bandits

Explore a research logbook and sync findings with an AI agent

Reproduction: Stochastic Weakly Convex Optimization Under Heavy-Tailed Noises

Explore research logbook and sync with coding agent

Reproduction: Robust Learnability of Sample-Compressible Distributions

Explore and share a logbook with an AI coding agent

Repro: Key Collision Attack on LLM Semantic Caching

Explore and sync a research logbook with your coding agent

Repro: Coupled Training with Privileged Information and Unlabeled Data

Explore and sync research logbook with an AI agent

Repro: Multi-Distribution Robust Conformal Prediction

Track and share experiment logs with AI collaboration

Repro: Riemannian Dueling Optimization

Track and organize research experiments in a collaborative logbook

Repro: Nonparametric LLM Evaluation from Preference Data

Repro: Clustering for directed graphs using parametrized random walk diffusion kernels

Explore experiment logs and sync with your coding agent

Repro: Local Hessian Spectral Filtering for Robust Intrinsic Dimension Estimation

Explore experiment logs and collaborate with a coding agent

Repro: Unfolded Laplacian Spectral Embedding

Browse and collaborate on experiment logbooks with an AI agent

Repro: Efficient Preference Poisoning Attack on Offline RLHF

Browse experiment logs and sync findings with an AI agent

Repro: R^2k is Theoretically Large Enough for Embedding-based Top-k Retrieval

Collaborate on research notes with an AI coding agent

Repro: Solving the Offline and Online Min-Max Problem of Non-smooth Submodular-Concave Functions

Explore and collaborate on research logbooks

Reproduction: Improved Convergence Analysis of Topology Dependence in Decentralized SGD

Explore and collaborate on research logbooks

Repro: Interventional Processes For Causal Uncertainty Quantification

Collaborate on a research logbook with an AI coding agent

Reproduction: Dynamic Regret via Discounted-to-Dynamic Reduction with Applications to Curved Losses and Adam Optimizer

Browse a research logbook and collaborate with your coding agent

Repro: Optimal Transport under Group Fairness Constraints

Explore experiment logs and collaborate with an AI agent

Reproduction: Minimax Optimal Strategy for Delayed Observations in Online Reinforcement Learning

Explore research logbook and sync with a coding agent

LR Annealing Reproduction — Overview

Explore project logs and collaborate with an AI agent

Repro: The (Marginal) Value of a Search Ad — Repeated Second-price Auctions

Explore and edit a project logbook with AI agent collaboration

Repro: Stochastic Linear Bandits with Parameter Noise

Browse experiment logs and sync findings with an AI agent

Reproduction: The Cost of Learning Under Multiple Change Points

Explore a research logbook and sync findings with an AI agent