Erik Schultheis

35 papers A* 9Journal 22Unranked 3
YearRankTypeTitle / Venue / Authors
2026 J jnl
DASH: Faster Shampoo via Batched Block Preconditioning and Efficient Inverse-Root Solvers.
CoRR
Ionut-Vlad Modoranu, Philip Zmushko, Erik Schultheis, Mher Safaryan, Dan Alistarh
2026 J jnl
Quartet II: Accurate LLM Pre-Training in NVFP4 by Improved Unbiased Gradient Estimation.
CoRR
Andrei Panferov, Erik Schultheis, Soroush Tabesh, Dan Alistarh
2025 J jnl
DynaSpec: Context-aware Dynamic Speculative Sampling for Large-Vocabulary Language Models.
CoRR
Jinbin Zhang, Nasib Ullah, Erik Schultheis, Rohit Babbar
2025 A* conf
ELMO : Efficiency via Low-precision and Peak Memory Optimization in Large Output Spaces.
ICML
Jinbin Zhang, Nasib Ullah, Erik Schultheis, Rohit Babbar
2025 J jnl
ELMO: Efficiency via Low-precision and Peak Memory Optimization in Large Output Spaces.
CoRR
Jinbin Zhang, Nasib Ullah, Erik Schultheis, Rohit Babbar
2025 — conf
How Well Calibrated are Extreme Multi-label Classifiers? An Empirical Analysis.
KDD (1)
Nasib Ullah, Erik Schultheis, Jinbin Zhang, Rohit Babbar
2025 J jnl
LLMQ: Efficient Lower-Precision Pretraining for Consumer GPUs.
CoRR
Erik Schultheis, Dan Alistarh
2025 J jnl
SVD-Free Low-Rank Adaptive Gradient Optimization for Large Language Models.
CoRR
Ionut-Vlad Modoranu, Mher Safaryan, Erik Schultheis, Dan Alistarh
2025 J jnl
Unbiased Loss Functions for Multilabel Classification with Missing Labels.
Trans. Mach. Learn. Res.
Erik Schultheis, Rohit Babbar
2024 A* conf
A General Online Algorithm for Optimizing Complex Performance Metrics.
ICML
Wojciech Kotlowski, Marek Wydmuch, Erik Schultheis, Rohit Babbar, Krzysztof Dembczynski
2024 J jnl
A General Online Algorithm for Optimizing Complex Performance Metrics.
CoRR
Wojciech Kotlowski, Marek Wydmuch, Erik Schultheis, Rohit Babbar, Krzysztof Dembczynski
2024 A* conf
Consistent algorithms for multi-label classification with macro-at-k metrics.
ICLR
Erik Schultheis, Wojciech Kotlowski, Marek Wydmuch, Rohit Babbar, Strom Borman, Krzysztof Dembczynski
2024 J jnl
Consistent algorithms for multi-label classification with macro-at-k metrics.
CoRR
Erik Schultheis, Wojciech Kotlowski, Marek Wydmuch, Rohit Babbar, Strom Borman, Krzysztof Dembczynski
2024 A* conf
Gandalf: Learning Label-label Correlations in Extreme Multi-label Classification via Label Features.
KDD
Siddhant Kharbanda, Devaansh Gupta, Erik Schultheis, Atmadeep Banerjee, Cho-Jui Hsieh, Rohit Babbar
2024 — conf
LLaMA-Annotate - Visualizing Token-Level Confidences for LLMs.
ECML/PKDD (8)
Erik Schultheis, St John
2024 J jnl
Labels in Extremes: How Well Calibrated are Extreme Multi-label Classifiers?
CoRR
Nasib Ullah, Erik Schultheis, Jinbin Zhang, Rohit Babbar
2024 J jnl
Learning label-label correlations in Extreme Multi-label Classification via Label Features.
CoRR
Siddhant Kharbanda, Devaansh Gupta, Erik Schultheis, Atmadeep Banerjee, Cho-Jui Hsieh, Rohit Babbar
2024 J jnl
Navigating Extremes: Dynamic Sparsity in Large Output Space.
CoRR
Nasib Ullah, Erik Schultheis, Mike Lasby, Yani Ioannou, Rohit Babbar
2024 A* conf
Navigating Extremes: Dynamic Sparsity in Large Output Spaces.
NeurIPS
Nasibullah Nasibullah, Erik Schultheis, Mike Lasby, Yani Ioannou, Rohit Babbar
2023 A* conf
Generalized test utilities for long-tail performance in extreme multi-label classification.
NeurIPS
Erik Schultheis, Marek Wydmuch, Wojciech Kotlowski, Rohit Babbar, Krzysztof Dembczynski
2023 J jnl
Generalized test utilities for long-tail performance in extreme multi-label classification.
CoRR
Erik Schultheis, Marek Wydmuch, Wojciech Kotlowski, Rohit Babbar, Krzysztof Dembczynski
2023 J jnl
Generating artificial displacement data of cracked specimen using physics-guided adversarial networks.
Mach. Learn. Sci. Technol.
David Melching, Erik Schultheis, Eric Breitbarth
2023 J jnl
Physics-guided adversarial networks for artificial digital image correlation data generation.
CoRR
David Melching, Erik Schultheis, Eric Breitbarth
2023 J jnl
Towards Memory-Efficient Training for Extremely Large Output Spaces - Learning with 500k Labels on a Single Commodity GPU.
CoRR
Erik Schultheis, Rohit Babbar
2023 — conf
Towards Memory-Efficient Training for Extremely Large Output Spaces - Learning with 670k Labels on a Single Commodity GPU.
ECML/PKDD (3)
Erik Schultheis, Rohit Babbar
2022 A* conf
CascadeXML: Rethinking Transformers for End-to-end Multi-resolution Training in Extreme Multi-label Classification.
NeurIPS
Siddhant Kharbanda, Atmadeep Banerjee, Erik Schultheis, Rohit Babbar
2022 J jnl
CascadeXML: Rethinking Transformers for End-to-end Multi-resolution Training in Extreme Multi-label Classification.
CoRR
Siddhant Kharbanda, Atmadeep Banerjee, Erik Schultheis, Rohit Babbar
2022 — ch.
Extreme Multicore Classification.
Mach. Learn. under Resour. Constraints Vol. 1 (1)
Erik Schultheis, Rohit Babbar
2022 A* conf
On Missing Labels, Long-tails and Propensities in Extreme Multi-label Classification.
KDD
Erik Schultheis, Marek Wydmuch, Rohit Babbar, Krzysztof Dembczynski
2022 J jnl
On Missing Labels, Long-tails and Propensities in Extreme Multi-label Classification.
CoRR
Erik Schultheis, Marek Wydmuch, Rohit Babbar, Krzysztof Dembczynski
2022 J jnl
Speeding-up one-versus-all training for extreme classification via mean-separating initialization.
Mach. Learn.
Erik Schultheis, Rohit Babbar
2021 A* conf
Convex Surrogates for Unbiased Loss Functions in Extreme Classification With Missing Labels.
WWW
Mohammadreza Qaraei, Erik Schultheis, Priyanshu Gupta, Rohit Babbar
2021 J jnl
Speeding-up One-vs-All Training for Extreme Classification via Smart Initialization.
CoRR
Erik Schultheis, Rohit Babbar
2021 J jnl
Unbiased Loss Functions for Multilabel Classification with Missing Labels.
CoRR
Erik Schultheis, Rohit Babbar
2020 J jnl
Unbiased Loss Functions for Extreme Classification With Missing Labels.
CoRR
Erik Schultheis, Mohammadreza Qaraei, Priyanshu Gupta, Rohit Babbar