Umong Sain

Umong

2

·

Umong51

AI & ML interests

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

Recent Activity

updated a Space 27 days ago

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

updated a bucket 27 days ago

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

published a bucket 27 days ago

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

View all activity

Organizations

Umong 's Spaces 34

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

Browse and collaborate on research logbooks

Repro - FlashOptim: Memory Efficient Optimizers for Large-Scale Training

Explore training logs and sync findings with an AI agent

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

Explore and organize experiment logs with AI collaboration

Repro - Training-Inference Consistent Segmented Execution for Long-Context LLMs

Explore and share LLM experiment logs

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

Explore simulation logs and collaborate with a coding agent

Repro - Conditional KRR: Injecting Unpenalized Features into Kernel Methods with Applications to Kernel Thresholding

Explore research logbook and sync AI‑generated findings

Repro - R²ᵏ is Theoretically Large Enough for Embedding-based Top-k Retrieval

Browse and collaborate on a project logbook with an AI agent

Repro - Less is More: Geometric Unlearning for LLMs with Minimal Data Disclosure

Explore and collaborate on a token‑efficient research logbook

FastMedExtract

MayerBhasha