Umong Sain PRO

Umong

2

·

Umong51

AI & ML interests

Natural Language Processing, Computer Vision, Contrastive Learning, Self-supervised Learning

Recent Activity

updated a Space about 21 hours ago

Umong/repro-tau-2-bench-evaluating-conversational-agents-in-a-dual-control-environment

updated a bucket about 21 hours ago

Umong/repro-tau-2-bench-evaluating-conversational-agents-in-a-dual-control-environment-artifacts

published a bucket about 21 hours ago

Umong/repro-tau-2-bench-evaluating-conversational-agents-in-a-dual-control-environment-artifacts

View all activity

Organizations

Umong 's Spaces 34

Reproduction: $\tau^2$-Bench: Evaluating Conversational Agents in a Dual-Control Environment

Collaborate with an AI agent to manage your project logbook

Reproduction: A Control-Theoretic View of Mamba on Stability and Robustness

Explore and collaborate on a research logbook via web interface

Reproduction: Inference from Quantized Data via Normal Variance-Mean Mixtures

Explore and sync research logbook with AI agent

Reproduction: Gradient Flow Through Diagram Expansions: Learning Regimes and Explicit Solutions

Explore and collaborate on research logbooks

Reproduction: Streaming Sliced Optimal Transport

Explore research logbook and collaborate with a coding agent

Reproduction: Alignment-Sensitive Minimax Rates for Spectral Algorithms with Learned Kernels

Explore research logbooks with interactive navigation

Reproduction: An Odd Estimator for Shapley Values

Collaborate on research logbook with your coding agent

Reproduction: A Linearly Convergent Proximal Subgradient Algorithm for Sparse Portfolio Optimization with Transaction Cost

Explore research logbook and collaborate with a coding agent

Reproduction: Probabilistic Bisection Algorithm Provably Achieves Exponential Convergence

Explore and share research logbooks with an AI collaborator

Reproduction: A Decision-Theoretic View of Test-Time Training: When, How Far, and Which Directions to Adapt

Explore and collaborate on research logbooks with an AI agent

Reproduction: Why Do We Need Warm-up? A Theoretical Perspective

Explore a research logbook and sync findings with an AI agent

Reproduction: Utility Boundary of Dataset Distillation: Scaling and Coverage Laws

Explore a project logbook and sync with a coding agent

Reproduction: FlexRank: Nested Low-Rank Knowledge Decomposition for Adaptive Model Deployment

Organize and sync research notes with an AI collaborator

Reproduction: d2: Improving Reasoning in Diffusion Language Models via Trajectory Likelihood Estimation

Track and collaborate on research logs with an AI assistant

Reproduction: Maximum Likelihood Reinforcement Learning

Explore experiment logs and sync findings with an AI agent

Reproduction: Training–Inference Consistent Segmented Execution for Long-Context LLMs

Explore experiment logs and collaborate with an AI agent

Reproduction: FOAM: Blocked State Folding for Memory-Efficient LLM Training

Explore and collaborate on research logbooks with an AI agent

Reproduction: Stable Deep Reinforcement Learning via Isotropic Gaussian Representations

Explore and collaborate on experiment logbooks online

Reproduction: FlashOptim: Optimizers for Memory-Efficient Training

Browse and record experiment logs with AI collaboration

Reproduction: Rex: A Family of Reversible Exponential (Stochastic) Runge-Kutta Solvers

Explore and collaborate on experiment logs in a web UI

Reproduction: Efficiently Training Time-to-First-Spike Spiking Neural Networks from Scratch

Explore experiment logs and collaborate with an AI agent

ETTFS reaches 92.38% on Fashion-MNIST

Explore and share experiment logs with your coding agent

SIGReg preserves plasticity on non-stationary CIFAR-10

Explore experiment logs in an interactive web logbook

Repro - Maximum Likelihood Reinforcement Learning

Explore experiment logs and sync with your coding agent