Amirkeivan Mohtashami

21 papers A* 5A 3Journal 13

Year	Rank	Type	Title / Venue / Authors
2025	A*	`conf`	CoTFormer: A Chain of Thought Driven Architecture with Budget-Adaptive Computation Cost at Inference. ICLR Amirkeivan Mohtashami, Matteo Pagliardini, Martin Jaggi
2024	A*	`conf`	DenseFormer: Enhancing Information Flow in Transformers via Depth Weighted Averaging. NeurIPS Matteo Pagliardini, Amirkeivan Mohtashami, François Fleuret, Martin Jaggi
2024	J	`jnl`	DenseFormer: Enhancing Information Flow in Transformers via Depth Weighted Averaging. CoRR Matteo Pagliardini, Amirkeivan Mohtashami, François Fleuret, Martin Jaggi
2024	A*	`conf`	QuaRot: Outlier-Free 4-Bit Inference in Rotated LLMs. NeurIPS Saleh Ashkboos, Amirkeivan Mohtashami, Maximilian L. Croci, Bo Li, Pashmina Cameron, Martin Jaggi, Dan Alistarh, Torsten Hoefler, James Hensman
2024	J	`jnl`	QuaRot: Outlier-Free 4-Bit Inference in Rotated LLMs. CoRR Saleh Ashkboos, Amirkeivan Mohtashami, Maximilian L. Croci, Bo Li, Martin Jaggi, Dan Alistarh, Torsten Hoefler, James Hensman
2023	J	`jnl`	CoTFormer: More Tokens With Attention Make Up For Less Depth. CoRR Amirkeivan Mohtashami, Matteo Pagliardini, Martin Jaggi
2023	J	`jnl`	Landmark Attention: Random-Access Infinite Context Length for Transformers. CoRR Amirkeivan Mohtashami, Martin Jaggi
2023	J	`jnl`	Learning Translation Quality Evaluation on Low Resource Languages from Large Language Models. CoRR Amirkeivan Mohtashami, Mauro Verzetti, Paul K. Rubenstein
2023	J	`jnl`	MEDITRON-70B: Scaling Medical Pretraining for Large Language Models. CoRR Zeming Chen, Alejandro Hernández-Cano, Angelika Romanou, Antoine Bonnet, Kyle Matoba, Francesco Salvi, Matteo Pagliardini, Simin Fan, Andreas Köpf, Amirkeivan Mohtashami, Alexandre Sallinen, Alireza Sakhaeirad, Vinitra Swamy, Igor Krawczuk, Deniz Bayazit, Axel Marmet, Syrielle Montariol, Mary-Anne Hartley, Martin Jaggi, Antoine Bosselut
2023	A*	`conf`	Random-Access Infinite Context Length for Transformers. NeurIPS Amirkeivan Mohtashami, Martin Jaggi
2023	J	`jnl`	Social Learning: Towards Collaborative Learning with Large Language Models. CoRR Amirkeivan Mohtashami, Florian Hartmann, Sian Gooding, Lukas Zilka, Matthew Sharifi, Blaise Agüera y Arcas
2023	A*	`conf`	Special Properties of Gradient Descent with Large Learning Rates. ICML Amirkeivan Mohtashami, Martin Jaggi, Sebastian U. Stich
2023	J	`jnl`	The splay-list: a distribution-adaptive concurrent skip-list. Distributed Comput. Vitaly Aksenov, Dan Alistarh, Alexandra Drozdova, Amirkeivan Mohtashami
2022	J	`jnl`	Characterizing & Finding Good Data Orderings for Fast Convergence of Sequential Gradient Methods. CoRR Amirkeivan Mohtashami, Sebastian U. Stich, Martin Jaggi
2022	A	`conf`	Masked Training of Neural Networks with Partial Gradients. AISTATS Amirkeivan Mohtashami, Martin Jaggi, Sebastian U. Stich
2022	J	`jnl`	On Avoiding Local Minima Using Gradient Descent With Large Learning Rates. CoRR Amirkeivan Mohtashami, Martin Jaggi, Sebastian U. Stich
2021	A	`conf`	Critical Parameters for Scalable Distributed Learning with Large Batches and Asynchronous Updates. AISTATS Sebastian U. Stich, Amirkeivan Mohtashami, Martin Jaggi
2021	J	`jnl`	Critical Parameters for Scalable Distributed Learning with Large Batches and Asynchronous Updates. CoRR Sebastian U. Stich, Amirkeivan Mohtashami, Martin Jaggi
2021	J	`jnl`	Simultaneous Training of Partially Masked Neural Networks. CoRR Amirkeivan Mohtashami, Martin Jaggi, Sebastian U. Stich
2020	A	`conf`	The Splay-List: A Distribution-Adaptive Concurrent Skip-List. DISC Vitaly Aksenov, Dan Alistarh, Alexandra Drozdova, Amirkeivan Mohtashami
2020	J	`jnl`	The Splay-List: A Distribution-Adaptive Concurrent Skip-List. CoRR Vitaly Aksenov, Dan Alistarh, Alexandra Drozdova, Amirkeivan Mohtashami