Enric Boix-Adserà

41 papers A* 10Journal 30

Year	Rank	Type	Title / Venue / Authors
2025	J	`jnl`	Aggregate and conquer: detecting and steering LLM concepts by combining nonlinear predictors over multiple layers. CoRR Daniel Beaglehole, Adityanarayanan Radhakrishnan, Enric Boix-Adserà, Mikhail Belkin
2025	J	`jnl`	FACT: the Features At Convergence Theorem for neural networks. CoRR Enric Boix-Adserà, Neil Mallinar, James B. Simon, Mikhail Belkin
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	J	`jnl`	On the inductive bias of infinite-depth ResNets and the bottleneck rank. CoRR Enric Boix-Adserà
2025	J	`jnl`	Secret mixtures of experts inside your LLM. CoRR Enric Boix-Adserà
2025	J	`jnl`	The Average-Case Complexity of Counting Cliques in Erdös-Rényi Hypergraphs. SIAM J. Comput. Enric Boix-Adserà, Matthew S. Brennan, Guy Bresler
2025	J	`jnl`	The power of fine-grained experts: Granularity boosts expressivity in Mixture of Experts. CoRR Enric Boix-Adserà, Philippe Rigollet
2024	—	`—`	Characterizations of how neural networks learn Enric Boix-Adserà
2024	A*	`conf`	Prompts have evil twins. EMNLP Rimon Melamed, Lucas H. McCabe, Tanay Wakhare, Yejin Kim, H. Howie Huang, Enric Boix-Adserà
2024	J	`jnl`	Towards a theory of model distillation. CoRR Enric Boix-Adserà
2024	A*	`conf`	When can transformers reason with abstract symbols? ICLR Enric Boix-Adserà, Omid Saremi, Emmanuel Abbe, Samy Bengio, Etai Littwin, Joshua M. Susskind
2023	J	`jnl`	PROPANE: Prompt design as an inverse problem. CoRR Rimon Melamed, Lucas H. McCabe, Tanay Wakhare, Yejin Kim, H. Howie Huang, Enric Boix-Adserà
2023	J	`jnl`	Polynomial-time algorithms for multimarginal optimal transport problems with structure. Math. Program. Jason M. Altschuler, Enric Boix-Adserà
2023	J	`jnl`	SGD learning on neural networks: leap complexity and saddle-to-saddle dynamics. CoRR Emmanuel Abbe, Enric Boix-Adserà, Theodor Misiakiewicz
2023	J	`jnl`	The NTK approximation is valid for longer than you think. CoRR Enric Boix-Adserà, Etai Littwin
2023	J	`jnl`	Tight conditions for when the NTK approximation is valid. Trans. Mach. Learn. Res. Enric Boix-Adserà, Etai Littwin
2023	A*	`conf`	Transformers learn through gradual rank increase. NeurIPS Emmanuel Abbe, Samy Bengio, Enric Boix-Adserà, Etai Littwin, Joshua M. Susskind
2023	J	`jnl`	Transformers learn through gradual rank increase. CoRR Enric Boix-Adserà, Etai Littwin, Emmanuel Abbe, Samy Bengio, Joshua M. Susskind
2023	J	`jnl`	When can transformers reason with abstract symbols? CoRR Enric Boix-Adserà, Omid Saremi, Emmanuel Abbe, Samy Bengio, Etai Littwin, Joshua M. Susskind
2022	A*	`conf`	GULP: a prediction-based metric between representations. NeurIPS Enric Boix-Adserà, Hannah Lawrence, George Stepaniants, Philippe Rigollet
2022	A*	`conf`	On the non-universality of deep learning: quantifying the cost of symmetry. NeurIPS Emmanuel Abbe, Enric Boix-Adserà
2022	J	`jnl`	On the non-universality of deep learning: quantifying the cost of symmetry. CoRR Emmanuel Abbe, Enric Boix-Adserà
2022	J	`jnl`	The merged-staircase property: a necessary and nearly sufficient condition for SGD learning of sparse functions on two-layer neural networks. CoRR Emmanuel Abbe, Enric Boix-Adserà, Theodor Misiakiewicz
2022	J	`jnl`	Wasserstein Barycenters Are NP-Hard to Compute. SIAM J. Math. Data Sci. Jason M. Altschuler, Enric Boix-Adserà
2021	A*	`conf`	Chow-Liu++: Optimal Prediction-Centric Learning of Tree Ising Models. FOCS Enric Boix-Adserà, Guy Bresler, Frederic Koehler
2021	J	`jnl`	Chow-Liu++: Optimal Prediction-Centric Learning of Tree Ising Models. CoRR Enric Boix-Adserà, Guy Bresler, Frederic Koehler
2021	J	`jnl`	Hardness results for Multimarginal Optimal Transport problems. Discret. Optim. Jason M. Altschuler, Enric Boix-Adserà
2021	J	`jnl`	The multiplayer Colonel Blotto game. Games Econ. Behav. Enric Boix-Adserà, Benjamin L. Edelman, Siddhartha Jayanti
2021	A*	`conf`	The staircase property: How hierarchical structure can guide deep learning. NeurIPS Emmanuel Abbe, Enric Boix-Adserà, Matthew S. Brennan, Guy Bresler, Dheeraj Nagaraj
2021	J	`jnl`	The staircase property: How hierarchical structure can guide deep learning. CoRR Emmanuel Abbe, Enric Boix-Adserà, Matthew S. Brennan, Guy Bresler, Dheeraj Nagaraj
2021	J	`jnl`	Wasserstein barycenters are NP-hard to compute. CoRR Jason M. Altschuler, Enric Boix-Adserà
2021	J	`jnl`	Wasserstein barycenters can be computed in polynomial time in fixed dimension. J. Mach. Learn. Res. Jason M. Altschuler, Enric Boix-Adserà
2020	J	`jnl`	Graph Powering and Spectral Robustness. SIAM J. Math. Data Sci. Emmanuel Abbe, Enric Boix-Adserà, Peter Ralli, Colin Sandon
2020	J	`jnl`	Hardness results for Multimarginal Optimal Transport problems. CoRR Jason M. Altschuler, Enric Boix-Adserà
2020	J	`jnl`	High-precision Wasserstein barycenters in polynomial time. CoRR Jason M. Altschuler, Enric Boix-Adserà
2020	J	`jnl`	Polynomial-time algorithms for Multimarginal Optimal Transport problems with structure. CoRR Jason M. Altschuler, Enric Boix-Adserà
2020	A*	`conf`	The Multiplayer Colonel Blotto Game. EC Enric Boix-Adserà, Benjamin L. Edelman, Siddhartha Jayanti
2020	J	`jnl`	The Multiplayer Colonel Blotto Game. CoRR Enric Boix-Adserà, Benjamin L. Edelman, Siddhartha Jayanti
2019	J	`jnl`	Approximating the Determinant of Well-Conditioned Matrices by Shallow Circuits. CoRR Enric Boix-Adserà, Lior Eldar, Saeed Mehraban
2019	A*	`conf`	Randomized Concurrent Set Union and Generalized Wake-Up. PODC Siddhartha V. Jayanti, Robert E. Tarjan, Enric Boix-Adserà
2019	A*	`conf`	The Average-Case Complexity of Counting Cliques in Erdős-Rényi Hypergraphs. FOCS Enric Boix-Adserà, Matthew S. Brennan, Guy Bresler