6 papers
Optimizing CUDA like a Human: Micro-Profiling Tools as Expert Surrogates for LLM-Based GPU Kernel Optimization
Jiading Gai, Shuai Zhang, Kaj Bostrom +6
We present KernelPro, a closed-loop multi-agent system that automatically generates, profiles, and iteratively optimizes GPU kernel code by integrating large language model (LLM) c…
Steering Large Reasoning Models towards Concise Reasoning via Flow Matching
Yawei Li, Benjamin Bergner, Yinghan Zhao +3
Large Reasoning Models (LRMs) excel at complex reasoning tasks, but their efficiency is often hampered by overly verbose outputs. Prior steering methods attempt to address this iss…
Retrieval-Augmented Decision Transformer: External Memory for In-context RL
Thomas Schmied, Fabian Paischer, Vihang Patil +3
In-context learning (ICL) is the ability of a model to learn a new task by observing a few exemplars in its context. While prevalent in NLP, this capability has recently also been…
A Large Recurrent Action Model: xLSTM enables Fast Inference for Robotics Tasks
Thomas Schmied, Thomas Adler, Vihang Patil +6
In recent years, there has been a trend in the field of Reinforcement Learning (RL) towards large action models trained offline on large-scale datasets via sequence modeling. Exist…
Simplified priors for Object-Centric Learning
Vihang Patil, Andreas Radler, Daniel Klotz +1
Humans excel at abstracting data and constructing \emph{reusable} concepts, a capability lacking in current continual learning systems. The field of object-centric learning address…
Contrastive Abstraction for Reinforcement Learning
Vihang Patil, Markus Hofmarcher, Elisabeth Rumetshofer +1
Learning agents with reinforcement learning is difficult when dealing with long trajectories that involve a large number of states. To address these learning problems effectively,…