Eran Malach - RankMe

73 papers A* 24Journal 45Unranked 4

Year	Rank	Type	Title / Venue / Authors
2025	A*	`conf`	A New Perspective on Shampoo's Preconditioner. ICLR Depen Morwani, Itai Shapira, Nikhil Vyas, Eran Malach, Sham M. Kakade, Lucas Janson
2025	J	`jnl`	A Taxonomy of Transcendence. CoRR Natalie Abreu, Edwin Zhang, Eran Malach, Naomi Saphra
2025	J	`jnl`	Decomposing Elements of Problem Solving: What "Math" Does RL Teach? CoRR Tian Qin, Core Francisco Park, Mujin Kwun, Aaron Walsman, Eran Malach, Nikhil Anand, Hidenori Tanaka, David Alvarez-Melis
2025	A*	`conf`	Don't stop me Now: Embedding based Scheduling for LLMS. ICLR Rana Shahout, Eran Malach, Chunwei Liu, Weifan Jiang, Minlan Yu, Michael Mitzenmacher
2025	J	`jnl`	Echo Chamber: RL Post-training Amplifies Behaviors Learned in Pretraining. CoRR Rosie Zhao, Alexandru Meterez, Sham M. Kakade, Cengiz Pehlevan, Samy Jelassi, Eran Malach
2025	J	`jnl`	How Reinforcement Learning After Next-Token Prediction Facilitates Learning. CoRR Nikolaos Tsilivis, Eran Malach, Karen Ullrich, Julia Kempe
2025	J	`jnl`	LLM-ERM: Sample-Efficient Program Learning via LLM-Guided Search. CoRR Shivam Singhal, Eran Malach, Tomaso A. Poggio, Tomer Galanti
2025	J	`jnl`	Let Me Think! A Long Chain-of-Thought Can Be Worth Exponentially Many Short Ones. CoRR Parsa Mirtaheri, Ezra Edelman, Samy Jelassi, Eran Malach, Enric Boix-Adserà
2025	—	`conf`	LoRA Soups: Merging LoRAs for Practical Skill Composition Tasks. COLING (Industry) Akshara Prabhakar, Yuanzhi Li, Karthik Narasimhan, Sham M. Kakade, Eran Malach, Samy Jelassi
2025	J	`jnl`	Loss-to-Loss Prediction: Scaling Laws for All Datasets. Trans. Mach. Learn. Res. David Brandfonbrener, Nikhil Anand, Nikhil Vyas, Eran Malach, Sham M. Kakade
2025	A*	`conf`	Mixture of Parrots: Experts improve memorization more than reasoning. ICLR Samy Jelassi, Clara Mohri, David Brandfonbrener, Alex Gu, Nikhil Vyas, Nikhil Anand, David Alvarez-Melis, Yuanzhi Li, Sham M. Kakade, Eran Malach
2025	J	`jnl`	Rethinking JEPA: Compute-Efficient Video SSL with Frozen Teachers. CoRR Xianhang Li, Chen Huang, Chun-Liang Li, Eran Malach, Josh M. Susskind, Vimal Thilak, Etai Littwin
2025	A*	`conf`	The Power of Random Features and the Limits of Distribution-Free Gradient Descent. ICML Ari Karchmer, Eran Malach
2025	J	`jnl`	The Power of Random Features and the Limits of Distribution-Free Gradient Descent. CoRR Ari Karchmer, Eran Malach
2025	A*	`conf`	The Role of Sparsity for Length Generalization in LLMs. ICML Noah Golowich, Samy Jelassi, David Brandfonbrener, Sham M. Kakade, Eran Malach
2025	J	`jnl`	The Role of Sparsity for Length Generalization in Transformers. CoRR Noah Golowich, Samy Jelassi, David Brandfonbrener, Sham M. Kakade, Eran Malach
2025	J	`jnl`	To Backtrack or Not to Backtrack: When Sequential Search Limits Model Reasoning. CoRR Tian Qin, David Alvarez-Melis, Samy Jelassi, Eran Malach
2025	J	`jnl`	To Infinity and Beyond: Tool-Use Unlocks Length Generalization in State Space Models. CoRR Eran Malach, Omid Saremi, Sinead Williamson, Arwen Bradley, Aryo Lotfi, Emmanuel Abbe, Joshua M. Susskind, Etai Littwin
2025	A*	`conf`	Universal Length Generalization with Turing Programs. ICML Kaiying Hou, David Brandfonbrener, Sham M. Kakade, Samy Jelassi, Eran Malach
2024	J	`jnl`	A New Perspective on Shampoo's Preconditioner. CoRR Depen Morwani, Itai Shapira, Nikhil Vyas, Eran Malach, Sham M. Kakade, Lucas Janson
2024	A*	`conf`	Auto-Regressive Next-Token Predictors are Universal Learners. ICML Eran Malach
2024	J	`jnl`	Don't Stop Me Now: Embedding Based Scheduling for LLMs. CoRR Rana Shahout, Eran Malach, Chunwei Liu, Weifan Jiang, Minlan Yu, Michael Mitzenmacher
2024	—	`conf`	Is ML-Based Cryptanalysis Inherently Limited? Simulating Cryptographic Adversaries via Gradient-Based Methods. CRYPTO (6) Avital Shafran, Eran Malach, Thomas Ristenpart, Gil Segev, Stefano Tessaro
2024	J	`jnl`	Is ML-Based Cryptanalysis Inherently Limited? Simulating Cryptographic Adversaries via Gradient-Based Methods. IACR Cryptol. ePrint Arch. Avital Shafran, Eran Malach, Thomas Ristenpart, Gil Segev, Stefano Tessaro
2024	J	`jnl`	LoRA Soups: Merging LoRAs for Practical Skill Composition Tasks. CoRR Akshara Prabhakar, Yuanzhi Li, Karthik Narasimhan, Sham M. Kakade, Eran Malach, Samy Jelassi
2024	J	`jnl`	Loss-to-Loss Prediction: Scaling Laws for All Datasets. CoRR David Brandfonbrener, Nikhil Anand, Nikhil Vyas, Eran Malach, Sham M. Kakade
2024	J	`jnl`	Mixture of Parrots: Experts improve memorization more than reasoning. CoRR Samy Jelassi, Clara Mohri, David Brandfonbrener, Alex Gu, Nikhil Vyas, Nikhil Anand, David Alvarez-Melis, Yuanzhi Li, Sham M. Kakade, Eran Malach
2024	A*	`conf`	On the Power of Decision Trees in Auto-Regressive Language Modeling. NeurIPS Yulu Gan, Tomer Galanti, Tomaso A. Poggio, Eran Malach
2024	J	`jnl`	On the Power of Decision Trees in Auto-Regressive Language Modeling. CoRR Yulu Gan, Tomer Galanti, Tomaso A. Poggio, Eran Malach
2024	A*	`conf`	Repeat After Me: Transformers are Better than State Space Models at Copying. ICML Samy Jelassi, David Brandfonbrener, Sham M. Kakade, Eran Malach
2024	J	`jnl`	Repeat After Me: Transformers are Better than State Space Models at Copying. CoRR Samy Jelassi, David Brandfonbrener, Sham M. Kakade, Eran Malach
2024	A*	`conf`	The Evolution of Statistical Induction Heads: In-Context Learning Markov Chains. NeurIPS Ezra Edelman, Nikolaos Tsilivis, Benjamin L. Edelman, Eran Malach, Surbhi Goel
2024	J	`jnl`	The Evolution of Statistical Induction Heads: In-Context Learning Markov Chains. CoRR Benjamin L. Edelman, Ezra Edelman, Surbhi Goel, Eran Malach, Nikolaos Tsilivis
2024	A*	`conf`	Transcendence: Generative Models Can Outperform The Experts That Train Them. NeurIPS Edwin Zhang, Vincent Zhu, Naomi Saphra, Anat Kleiman, Benjamin L. Edelman, Milind Tambe, Sham M. Kakade, Eran Malach
2024	J	`jnl`	Transcendence: Generative Models Can Outperform The Experts That Train Them. CoRR Edwin Zhang, Vincent Zhu, Naomi Saphra, Anat Kleiman, Benjamin L. Edelman, Milind Tambe, Sham M. Kakade, Eran Malach
2024	J	`jnl`	Universal Length Generalization with Turing Programs. CoRR Kaiying Hou, David Brandfonbrener, Sham M. Kakade, Samy Jelassi, Eran Malach
2023	J	`jnl`	Auto-Regressive Next-Token Predictors are Universal Learners. CoRR Eran Malach
2023	J	`jnl`	Corgi^2: A Hybrid Offline-Online Approach To Storage-Aware Data Shuffling For SGD. CoRR Etay Livne, Gal Kaplun, Eran Malach, Shai Shalev-Schwatz
2023	A*	`conf`	Pareto Frontiers in Deep Feature Learning: Data, Compute, Width, and Luck. NeurIPS Benjamin L. Edelman, Surbhi Goel, Sham M. Kakade, Eran Malach, Cyril Zhang
2023	J	`jnl`	Pareto Frontiers in Neural Feature Learning: Data, Compute, Width, and Luck. CoRR Benjamin L. Edelman, Surbhi Goel, Sham M. Kakade, Eran Malach, Cyril Zhang
2023	J	`jnl`	SubTuning: Efficient Finetuning for Multi-Task Learning. CoRR Gal Kaplun, Andrey Gurevich, Tal Swisa, Mazor David, Shai Shalev-Shwartz, Eran Malach
2022	A*	`conf`	Efficient Learning of CNNs using Patch Based Features. ICML Alon Brutzkus, Amir Globerson, Eran Malach, Alon Regev Netser, Shai Shalev-Shwartz
2022	A*	`conf`	Hidden Progress in Deep Learning: SGD Learns Parities Near the Computational Limit. NeurIPS Boaz Barak, Benjamin L. Edelman, Surbhi Goel, Sham M. Kakade, Eran Malach, Cyril Zhang
2022	J	`jnl`	Hidden Progress in Deep Learning: SGD Learns Parities Near the Computational Limit. CoRR Boaz Barak, Benjamin L. Edelman, Surbhi Goel, Sham M. Kakade, Eran Malach, Cyril Zhang
2022	A*	`conf`	Knowledge Distillation: Bad Models Can Be Good Role Models. NeurIPS Gal Kaplun, Eran Malach, Preetum Nakkiran, Shai Shalev-Shwartz
2022	J	`jnl`	Knowledge Distillation: Bad Models Can Be Good Role Models. CoRR Gal Kaplun, Eran Malach, Preetum Nakkiran, Shai Shalev-Shwartz
2022	J	`jnl`	When Hardness of Approximation Meets Hardness of Learning. J. Mach. Learn. Res. Eran Malach, Shai Shalev-Shwartz
2021	A*	`conf`	Computational Separation Between Convolutional and Fully-Connected Networks. ICLR Eran Malach, Shai Shalev-Shwartz
2021	A*	`conf`	On the Power of Differentiable Learning versus PAC and SQ Learning. NeurIPS Emmanuel Abbe, Pritish Kamath, Eran Malach, Colin Sandon, Nathan Srebro
2021	J	`jnl`	On the Power of Differentiable Learning versus PAC and SQ Learning. CoRR Emmanuel Abbe, Pritish Kamath, Eran Malach, Colin Sandon, Nathan Srebro
2021	A*	`conf`	Quantifying the Benefit of Using Differentiable Learning over Tangent Kernels. ICML Eran Malach, Pritish Kamath, Emmanuel Abbe, Nathan Srebro
2021	J	`jnl`	Quantifying the Benefit of Using Differentiable Learning over Tangent Kernels. CoRR Eran Malach, Pritish Kamath, Emmanuel Abbe, Nathan Srebro
2021	A*	`conf`	The Connection Between Approximation, Depth Separation and Learnability in Neural Networks. COLT Eran Malach, Gilad Yehudai, Shai Shalev-Shwartz, Ohad Shamir
2021	J	`jnl`	The Connection Between Approximation, Depth Separation and Learnability in Neural Networks. CoRR Eran Malach, Gilad Yehudai, Shai Shalev-Shwartz, Ohad Shamir
2020	J	`jnl`	Computational Separation Between Convolutional and Fully-Connected Networks. CoRR Eran Malach, Shai Shalev-Shwartz
2020	A*	`conf`	ID3 Learns Juntas for Smoothed Product Distributions. COLT Alon Brutzkus, Amit Daniely, Eran Malach
2020	A*	`conf`	Learning Parities with Neural Networks. NeurIPS Amit Daniely, Eran Malach
2020	J	`jnl`	Learning Parities with Neural Networks. CoRR Amit Daniely, Eran Malach
2020	A*	`conf`	Proving the Lottery Ticket Hypothesis: Pruning is All You Need. ICML Eran Malach, Gilad Yehudai, Shai Shalev-Shwartz, Ohad Shamir
2020	J	`jnl`	Proving the Lottery Ticket Hypothesis: Pruning is All You Need. CoRR Eran Malach, Gilad Yehudai, Shai Shalev-Shwartz, Ohad Shamir
2020	A*	`conf`	The Implications of Local Correlation on Learning Some Deep Functions. NeurIPS Eran Malach, Shai Shalev-Shwartz
2020	J	`jnl`	When Hardness of Approximation Meets Hardness of Learning. CoRR Eran Malach, Shai Shalev-Shwartz
2019	J	`jnl`	Decoupling Gating from Linearity. CoRR Jonathan Fiat, Eran Malach, Shai Shalev-Shwartz
2019	J	`jnl`	ID3 Learns Juntas for Smoothed Product Distributions. CoRR Alon Brutzkus, Amit Daniely, Eran Malach
2019	A*	`conf`	Is Deeper Better only when Shallow is Good? NeurIPS Eran Malach, Shai Shalev-Shwartz
2019	J	`jnl`	Is Deeper Better only when Shallow is Good? CoRR Eran Malach, Shai Shalev-Shwartz
2019	J	`jnl`	Learning Boolean Circuits with Neural Networks. CoRR Eran Malach, Shai Shalev-Shwartz
2019	J	`jnl`	On the Optimality of Trees Generated by ID3. CoRR Alon Brutzkus, Amit Daniely, Eran Malach
2018	J	`jnl`	A Provably Correct Algorithm for Deep Learning that Actually Works. CoRR Eran Malach, Shai Shalev-Shwartz
2018	—	`conf`	SGD Learns Over-parameterized Networks that Provably Generalize on Linearly Separable Data. ICLR (Poster) Alon Brutzkus, Amir Globerson, Eran Malach, Shai Shalev-Shwartz
2017	—	`conf`	Decoupling "when to update" from "how to update". NIPS Eran Malach, Shai Shalev-Shwartz
2017	J	`jnl`	Decoupling "when to update" from "how to update". CoRR Eran Malach, Shai Shalev-Shwartz
2017	J	`jnl`	SGD Learns Over-parameterized Networks that Provably Generalize on Linearly Separable Data. CoRR Alon Brutzkus, Amir Globerson, Eran Malach, Shai Shalev-Shwartz