Razvan Pascanu
~Razvan_Pascanu1
27
论文总数
13.5
年均投稿
平均评分
接收情况17/27
会议分布
ICLR
14
NeurIPS
8
COLM
3
ICML
2
发表论文 (27 篇)
202516 篇
4
RAT: Bridging RNN Efficiency and Attention Accuracy via Chunk-based Sequence Modeling
NeurIPS 2025Poster
4
Meta-learning how to Share Credit among Macro-Actions
NeurIPS 2025Poster
5
softmax is not enough (for sharp out-of-distribution)
ICLR 2025Rejected
5
Round and Round We Go! What makes Rotary Positional Encodings useful?
ICLR 2025Poster
4
Is multitask learning all you need in continual learning?
ICLR 2025Rejected
4
Softmax is not Enough (for Sharp Size Generalisation)
ICML 2025Poster
4
MS-SSM: A Multi-Scale State Space Model for Efficient Sequence Modeling
COLM 2025Poster
5
Attention as a Hypernetwork
ICLR 2025Oral
3
Retrieval-Augmented Decision Transformer: External Memory for In-context RL
ICLR 2025Rejected
4
How do language models learn facts? Dynamics, curricula and hallucinations
COLM 2025Poster
3
Torque-Aware Momentum
ICLR 2025Rejected
4
Why do LLMs attend to the first token?
COLM 2025Poster
4
Transformers meet Neural Algorithmic Reasoners
ICLR 2025Rejected
4
A Large Recurrent Action Model: xLSTM enables Fast Inference for Robotics Tasks
ICML 2025Poster
4
A Large Recurrent Action Model: xLSTM enables Fast Inference for Robotics Tasks
ICLR 2025Rejected
4
Plasticity as the Mirror of Empowerment
NeurIPS 2025Spotlight
202411 篇
3
Building on Efficient Foundations: Effective Training of LLMs with Structured Feedforward Layers
NeurIPS 2024Poster
4
Kalman Filter for Online Classification of Non-Stationary Data
ICLR 2024Poster
4
Towards Perpetually Trainable Neural Networks
ICLR 2024Rejected
4
A Demon at Work: Leveraging Neuron Death for Efficient Neural Network Pruning
ICLR 2024Rejected
6
No Representation, No Trust: Connecting Representation, Collapse, and Trust Issues in PPO
NeurIPS 2024Poster
4
Non-Stationary Learning of Neural Networks with Automatic Soft Parameter Reset
NeurIPS 2024Poster
4
The Role of Forgetting in Fine-Tuning Reinforcement Learning Models
ICLR 2024Rejected
4
Normalization and effective learning rates in reinforcement learning
NeurIPS 2024Poster
3
Transformers need glasses! Information over-squashing in language tasks
NeurIPS 2024Poster
4
Promoting Exploration in Memory-Augmented Adam using Critical Momenta
ICLR 2024Rejected
4
Discovering modular solutions that generalize compositionally
ICLR 2024Poster