Srishti Saha

Srishti280992

·

AI & ML interests

None yet

Recent Activity

upvoted an article about 2 months ago

What We Learned by Reproducing 2,200 papers from ICML

new activity 2 months ago

ICML-2026-agent-repro/challenge:Old logbooks suddenly changed to 'queued' after the deadline ended - were not uploaded/changed from my end!

updated a Space 2 months ago

Srishti280992/repro-deep-networks-learn-to-parse-uniform-depth-context-free-languages-from-local-statistics

View all activity

Organizations

Srishti280992 's Spaces 272

Reproduction: Statistical Early Stopping for Reasoning Models

Explore and manage experiment logs, traces, and workspace files

Reproduction: Accuracy-First Rényi Differential Privacy and Post-Processing Immunity

Explore code logs, traces, and workspace with collaborative AI

Reproduction: Let the Prototype Guide You: Robust Aggregation of Sparse Multi-Class Annotations via Annotator Prototype Learning

Browse project logs, traces, and workspace in a web UI

Reproduction: Alignment between Brains and AI: Evidence for Convergent Evolution across Modalities, Scales and Training Trajectories

Explore and manage project logs with AI collaboration

Reproduction: Causal Matrix Completion under Multiple Treatments via Mixed Synthetic Nearest Neighbors

Explore project logs, code, and traces via an interactive web UI

Reproduction: MemoryBench: A Benchmark for Memory and Continual Learning in LLM Systems

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

Reproduction: Time series saliency maps: Explaining models across multiple domains

Explore and collaborate on project logs with an AI agent

Reproduction: Revisiting Zeroth-Order Hessian Approximation: A Single-Step Policy Optimization Lens

Explore a project logbook with code, traces, and AI collaboration

Reproduction: A Capacity-Based Rationale for Multi-Head Attention

Explore experiment logs, traces, and workspace in one view

Reproduction: Why Agentic Theorem Prover Works: A Statistical Provability Theory of Mathematical Reasoning Models

Explore code logs, traces, and workspace with agent collaboration

Reproduction: Bregman meets Levy: Stochastic Mirror Descent with Heavy-Tailed Noise in Continuous and Discrete Time

Explore and manage experiment logs, traces, and workspace

Reproduction: Theoretical Investigation on Inductive Bias of Isolation Forest

Explore code logs, traces, and workspace with agent collaboration

Reproduction: (Doubly) Exponential Lower Bounds for Follow the Regularized Leader in Potential Games

Explore code, traces, and workspace for research logbooks

Reproduction: Reinforcement Learning with Verifiable Rewards: GRPO's Loss, Dynamics, and Success Amplification

Explore project logs, code, and traces in an interactive logbook

Reproduction: Asymptotically Optimal Sequential Testing with Markovian Data

Explore project logs, code, and execution traces in a web logbook

Reproduction: Best-of-Both-Worlds for Heavy-Tailed Markov Decision Processes

Explore project logs, traces, and workspace in a web interface

Reproduction: Momentum Further Constrains Sharpness at the Edge of Stochastic Stability

Explore and manage experiment logs with a collaborative web UI

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

Explore code, traces, and workspace in an interactive logbook

Reproduction: HaloProbe: Bayesian Detection and Mitigation of Object Hallucinations in Vision-Language Models

Explore project logs and collaborate with an AI agent

Reproduction: Continuous Variable Hamiltonian Learning at Heisenberg Limit via Displacement-Random Unitary Transformation

Explore project logs, traces, and workspace in a web dashboard

Reproduction: Efficient privacy loss accounting for subsampling and random allocation

Explore code logs, execution traces, and workspace in one view

Reproduction: Perceptrons and Localization of Attention's Mean-Field Landscape

View and organize experiment logs, traces, and workspace

Reproduction: Planar Symmetric Pattern Generation

Explore code, traces, and workspace in a collaborative logbook

Reproduction: All ERMs Can Fail in Stochastic Convex Optimization Lower Bounds in Linear Dimension

Explore code, traces, and workspace in an interactive logbook