Hong Seong Pyo

SabaPivot

·

AI & ML interests

AI Safety

Recent Activity

updated a Space about 6 hours ago

SabaPivot/repro-a-formal-comparison-between-chain-of-thought-and-latent-thought

updated a Space about 6 hours ago

SabaPivot/repro-facts-in-stats-impacts-of-pretraining-diversity-on-language-model-generalization

updated a Space about 7 hours ago

SabaPivot/repro-dissecting-multimodal-in-context-learning-modality-asymmetries-and-circuit-dynamics-in-mod

View all activity

Organizations

SabaPivot 's Spaces 321

Reproduction: Gaussian Mean Field Variational Inference can Overestimate Predictive Variance

Explore research logbooks and collaborate with an AI agent

Reproduction: Can Adaptive Gradient Methods Converge under Heavy-Tailed Noise? A Case Study of AdaGrad

Browse code logs, traces, and workspace files

Reproduction: Mirror Descent Under Generalized Smoothness

Explore code, traces, and workspace with an interactive logbook

Reproduction: Online Change Point Detection for Multivariate Inhomogeneous Poisson Processes Time Series

Browse and collaborate on a project logbook online

Reproduction: Quantifying LLM Attention-Head Stability: Implications for Circuit Universality

Collaborate with your coding agent in a shared logbook

Reproduction: Anytime Detection of Strategic Deviations in Multi-Agent Systems

Explore experiment logs and collaborate with an AI agent

Reproduction: Row-Stochastic Matrices Can Provably Outperform Doubly Stochastic Matrices in Decentralized Learning

Explore code logs, traces, and workspace in a web logbook

Reproduction: An Odd Estimator for Shapley Values

Track research logs and sync updates with your AI agent

Reproduction: Unveiling the Structure of Do-Calculus Reasoning via Derivation Graphs

Explore and collaborate on structured research logbooks

Reproduction: Non-Uniform Noise-to-Signal Ratio in the REINFORCE Policy-Gradient Estimator

Explore research logbooks and collaborate with an AI agent

Reproduction: Discretized Density-Guided Source-Free Domain Adaptation for Regression

Explore experiment logs and sync with AI agent

Reproduction: Bridging Time and Frequency: A Joint Modeling Framework for Irregular Multivariate Time Series Forecasting

Log and collaborate on research notes with an AI agent

Repro - To Grok Grokking: Provable Grokking in Ridge Regression

Explore experiment logbooks and sync with your AI agent

Repro - Who Said Neural Networks Aren't Linear?

Explore project logbook and sync with your coding agent

Repro - Success-Conditioning as Policy Improvement: The Optimization Problem Solved by Imitating Success

Collaborate with your AI agent to view and update a project logbook

Repro - Softmax as Linear Attention in the Large-Prompt Regime

Track research notes and sync with AI agent

Repro - On Structured State-Space Duality

View and sync a structured logbook with an AI agent

Repro - Understanding Generalization and Forgetting in In-Context Continual Learning

Browse experiment logbook and collaborate with an AI agent

Repro - Flat Minima and Generalization: Insights from Stochastic Convex Optimization

Explore and sync research logbook with your coding agent

Reproduction: A Tight Theory of Error Feedback Algorithms in Distributed Optimization

Explore research logbook and sync with your AI agent

Reproduction: Finding Most Influential Sets

Log and share experiment findings with an AI collaborator

Reproduction: HiPPO Zoo: Explicit Memory Mechanisms for Interpretable State Space Models

Explore experiment logs and sync with your coding agent

Reproduction: Balancing Learning Rates Across Layers: Exact Two-Step Dynamics and Optimal Scaling in Linear Neural Networks

Browse a research logbook and collaborate with a coding agent

Reproduction: Optimal Unconstrained Self-Distillation in Ridge Regression: Strict Improvements, Precise Asymptotics, and One-Shot Tuning

Explore and sync experiment logs with a coding agent