Reproducible ML Research & Open Science

vimarsh/repro-rational-transductors

vimarsh/repro-federated-bilevel-performative-prediction

DineshAI/nf7JT1jCSy

vimarsh/repro-adversarially-robust-approximate-furthest-neighbor

vimarsh/repro-leveraging-gauge-freedom-for-non-gradient-population-dynamics

vimarsh/repro-combinatorial-sparse-pca-beyond-the-spiked-identity-model

vimarsh/repro-connecting-independently-trained-modes-layer-wise-low-loss-paths

vimarsh/repro-deep-networks-learn-deep-hierarchical-models

vimarsh/repro-fedfit-federated-dynamic-sparse-training-via-fisher-information-scoring

vimarsh/repro-stochastic-sparse-attention-for-memory-bound-inference

vimarsh/repro-delving-into-muon-and-beyond-deep-analysis-and-extensions

vimarsh/Repro-SlideSparse-ICML-33765

vimarsh/repro-a-capacity-based-rationale-for-multi-head-attention

vimarsh/repro-sample-margin-aware-recalibration-of-temperature-scaling

vimarsh/repro-on-the-theory-of-continual-learning-with-gradient-descent-for-neural-networks

vimarsh/repro-mimomamba-matrix-valued-attention

vimarsh/repro-how-many-different-outputs-can-a-transformer-generate

vimarsh/repro-how-can-mamba-learn-in-context-with-outliers-and-generalize-provably

vimarsh/repro-little-by-little-continual-learning-via-incremental-mixture-of-rank-1-associative-memory-e

vimarsh/repro-adaptive-personalized-fl-multi-task-averaging-kernel-mean-embedding

vimarsh/repro-single-head-attention-inductive-bias-spectral-highdim

vimarsh/repro-why-linear-rnns-more-parallelizable

vimarsh/repro-event2vec-neuromorphic-events-vector-space

vimarsh/repro-token-sparse-attention-long-context-interleaved-selection

vimarsh/repro-rmnp-row-momentum-normalized-preconditioning-for-scalable-matrix-based-optimization

vimarsh/icml2026-repro-32912-efficient-landing

vimarsh/repro-full-batch-gd-outperforms-one-pass-sgd-sample-complexity-separation

vimarsh/repro-quantifying-the-noise-sensitivity-of-the-wasserstein-metric-for-images

vimarsh/repro-turning-stale-gradients-into-stable-gradients-coherent-coordinate-descent

vimarsh/repro-float8-2bits-entropy-coding-enables-data-free-model-compression

vimarsh/repro-revit-roto-reflection-equivariant-convolutional-vision-transformer

vimarsh/repro-understanding-dynamics-of-adam-in-zero-sum-games-an-ode-approach

vimarsh/repro-toast-token-channel-selection-and-structured-pruning-for-efficient-vit

vimarsh/repro-costaware-stopping-bayesopt

vimarsh/repro-any3d-vla-enhancing-vla-robustness-via-diverse-point-clouds

ICML-2026-agent-repro/challenge