Dimitris Papailiopoulos

Everything Everywhere All at Once: LLMs can In-Context Learn Multiple Tasks in Superposition ICML 2025 From Artificial Needles to Real Haystacks: Improving Retrieval Capabilities in LLMs by Finetuning on Synthetic Data ICLR 2025 How Well Can Transformers Emulate In-Context Newton’s Method? AISTATS 2025 Lexico: Extreme KV Cache Compression via Sparse Coding over Universal Dictionaries ICML 2025 Self-Improving Transformers Overcome Easy-to-Hard and Length Generalization Challenges ICML 2025 VersaPRM: Multi-Domain Process Reward Model via Synthetic Reasoning Data ICML 2025 Can Mamba Learn How To Learn? A Comparative Study on In-Context Learning Tasks ICML 2024 CHAI: Clustered Head Attention for Efficient LLM Inference ICML 2024 Looped Transformers are Better at Learning Learning Algorithms ICLR 2024 Teaching Arithmetic to Small Transformers ICLR 2024 Dissecting Chain-of-Thought: Compositionality through In-Context Filtering and Learning NeurIPS 2023 Looped Transformers as Programmable Computers ICML 2023 Prompted LLMs as Chatbot Modules for Long Open-domain Conversation ACL 2023 The Expressive Power of Tuning Only the Normalization Layers COLT 2023 Transformers as Algorithms: Generalization and Stability in In-context Learning ICML 2023 Finding Nearly Everything within Random Binary Networks AISTATS 2022 GenLabel: Mixup Relabeling using Generative Models ICML 2022 LIFT: Language-Interfaced Fine-Tuning for Non-language Machine Learning Tasks NeurIPS 2022 Permutation-Based SGD: Is Random Optimal? ICLR 2022 Rare Gems: Finding Lottery Tickets at Initialization NeurIPS 2022 Utilizing Language-Image Pretraining for Efficient and Robust Bilingual Word Alignment EMNLP 2022 An Exponential Improvement on the Memorization Capacity of Deep Threshold Networks NeurIPS 2021 Attack of the Tails: Yes, You Really Can Backdoor Federated Learning NeurIPS 2020 Bad Global Minima Exist and SGD Can Reach Them NeurIPS 2020 Closing the convergence gap of SGD without replacement ICML 2020 Federated Learning with Matched Averaging ICLR 2020 Optimal Lottery Tickets via Subset Sum: Logarithmic Over-Parameterization is Sufficient NeurIPS 2020 A Geometric Perspective on the Transferability of Adversarial Directions AISTATS 2019 DETOX: A Redundancy-based Framework for Faster and More Robust Gradient Aggregation NeurIPS 2019 Does Data Augmentation Lead to Positive Margin? ICML 2019 ATOMO: Communication-efficient Learning via Atomic Sparsification NeurIPS 2018 DRACO: Byzantine-resilient Distributed Training via Redundant Gradients ICML 2018 Gradient Diversity: a Key Ingredient for Scalable Distributed Learning AISTATS 2018 Stability and Generalization of Learning Algorithms that Converge to Global Optima ICML 2018 The Effect of Network Width on the Performance of Large-batch Training NeurIPS 2018 Bipartite Correlation Clustering: Maximizing Agreements AISTATS 2016 Cyclades: Conflict-free Asynchronous Machine Learning NeurIPS 2016 Orthogonal NMF through Subspace Exploration NeurIPS 2015 Parallel Correlation Clustering on Big Graphs NeurIPS 2015 Sparse PCA via Bipartite Matchings NeurIPS 2015 Finding Dense Subgraphs via Low-Rank Bilinear Optimization ICML 2014 Nonnegative Sparse PCA with Provable Guarantees ICML 2014 Sparse PCA through Low-rank Approximations ICML 2013