PaperHub

Razvan Pascanu

~Razvan_Pascanu1

27
论文总数
13.5
年均投稿
6.0
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

6.0
4

RAT: Bridging RNN Efficiency and Attention Accuracy via Chunk-based Sequence Modeling

NeurIPS 2025Poster
7.3
4

Meta-learning how to Share Credit among Macro-Actions

NeurIPS 2025Poster
6.2
5

softmax is not enough (for sharp out-of-distribution)

ICLR 2025Rejected
6.2
5

Round and Round We Go! What makes Rotary Positional Encodings useful?

ICLR 2025Poster
5.8
4

Is multitask learning all you need in continual learning?

ICLR 2025Rejected
6.1
4

Softmax is not Enough (for Sharp Size Generalisation)

ICML 2025Poster
6.8
4

MS-SSM: A Multi-Scale State Space Model for Efficient Sequence Modeling

COLM 2025Poster
7.6
5

Attention as a Hypernetwork

ICLR 2025Oral
4.7
3

Retrieval-Augmented Decision Transformer: External Memory for In-context RL

ICLR 2025Rejected
7.0
4

How do language models learn facts? Dynamics, curricula and hallucinations

COLM 2025Poster
4.7
3

Torque-Aware Momentum

ICLR 2025Rejected
6.8
4

Why do LLMs attend to the first token?

COLM 2025Poster
5.0
4

Transformers meet Neural Algorithmic Reasoners

ICLR 2025Rejected
6.1
4

A Large Recurrent Action Model: xLSTM enables Fast Inference for Robotics Tasks

ICML 2025Poster
5.8
4

A Large Recurrent Action Model: xLSTM enables Fast Inference for Robotics Tasks

ICLR 2025Rejected
7.3
4

Plasticity as the Mirror of Empowerment

NeurIPS 2025Spotlight

202411 篇

5.7
3

Building on Efficient Foundations: Effective Training of LLMs with Structured Feedforward Layers

NeurIPS 2024Poster
6.5
4

Kalman Filter for Online Classification of Non-Stationary Data

ICLR 2024Poster
5.8
4

Towards Perpetually Trainable Neural Networks

ICLR 2024Rejected
5.5
4

A Demon at Work: Leveraging Neuron Death for Efficient Neural Network Pruning

ICLR 2024Rejected
5.5
6

No Representation, No Trust: Connecting Representation, Collapse, and Trust Issues in PPO

NeurIPS 2024Poster
5.0
4

Non-Stationary Learning of Neural Networks with Automatic Soft Parameter Reset

NeurIPS 2024Poster
5.0
4

The Role of Forgetting in Fine-Tuning Reinforcement Learning Models

ICLR 2024Rejected
6.5
4

Normalization and effective learning rates in reinforcement learning

NeurIPS 2024Poster
6.3
3

Transformers need glasses! Information over-squashing in language tasks

NeurIPS 2024Poster
4.8
4

Promoting Exploration in Memory-Augmented Adam using Critical Momenta

ICLR 2024Rejected
6.5
4

Discovering modular solutions that generalize compositionally

ICLR 2024Poster
合作者 (20)