4 citations · 8 across the 14 of their papers we have counts for
17 papers
Task Vector Descent: Learning from Non-IID Batches
Anton Baumann, Jonas Hübotter, Zeynep Akata +1
A central challenge in continual learning is to acquire new knowledge without forgetting what the model has already learned. This challenge appears in language model training when…
Deep Epistemic Value Functions for Optimistic Exploration
Leander Diaz-Bone, Marco Bagatella, Jonas Hübotter +1
Principled exploration in reinforcement learning requires an agent to quantify its epistemic uncertainty and act to resolve it. Uncertainty over the value function provides a natur…
On Repulsive and Attractive Teachers: Separating Correctness from Behavior in Self-Distillation
Anton Baumann, Akmal Ashirmatov, Leo Schmidt-Traub +4
On-policy self-distillation provides dense, token-level supervision by conditioning a model on privileged information and distilling the resulting teacher distribution back into th…
Aligning Language Models from User Interactions
Thomas Kleine Buening, Jonas Hübotter, Barna Pásztor +3
Multi-turn user interactions are among the most abundant data produced by language models, yet we lack effective methods to learn from them. While typically discarded, these intera…
Self-Distillation Enables Continual Learning
Idan Shenfeld, Mehul Damani, Jonas Hübotter +1
Continual learning, enabling models to acquire new skills and knowledge without degrading existing capabilities, remains a fundamental challenge for foundation models. While on-pol…
Maximizing Prefix-Confidence at Test-Time Efficiently Improves Mathematical Reasoning
Matthias Otth, Jonas Hübotter, Ido Hakimi +1
Recent work has shown that language models can self-improve by maximizing their own confidence in their predictions, without relying on external verifiers or reward signals. In thi…
Test-time Offline Reinforcement Learning on Goal-related Experience
Marco Bagatella, Mert Albaba, Jonas Hübotter +2
Foundation models compress a large amount of information in a single, large neural network, which can then be queried for individual tasks. There are strong parallels between this…
Local Mixtures of Experts: Essentially Free Test-Time Training via Model Merging
Ryo Bertolissi, Jonas Hübotter, Ido Hakimi +1
Mixture of expert (MoE) models are a promising approach to increasing model capacity without increasing inference cost, and are core components of many state-of-the-art language mo…
Probabilistic Artificial Intelligence
Andreas Krause, Jonas Hübotter
Artificial intelligence commonly refers to the science and engineering of artificial systems that can carry out tasks generally associated with requiring aspects of human intellige…
LITE: Efficiently Estimating Gaussian Probability of Maximality
Nicolas Menet, Jonas Hübotter, Parnian Kassraie +1
We consider the problem of computing the probability of maximality (PoM) of a Gaussian random vector, i.e., the probability for each dimension to be maximal. This is a key challeng…
Efficiently Learning at Test-Time: Active Fine-Tuning of LLMs
Jonas Hübotter, Sascha Bongni, Ido Hakimi +1
Recent efforts in fine-tuning language models often rely on automatic data selection, commonly using Nearest Neighbors retrieval from large datasets. However, we theoretically show…
Active Fine-Tuning of Multi-Task Policies
Marco Bagatella, Jonas Hübotter, Georg Martius +1
Pre-trained generalist policies are rapidly gaining relevance in robot learning due to their promise of fast adaptation to novel, in-domain tasks. This adaptation often relies on c…
Transductive Active Learning: Theory and Applications
Jonas Hübotter, Bhavya Sukhija, Lenart Treven +2
We study a generalization of classical active learning to real-world settings with concrete prediction targets where sampling is restricted to an accessible region of the domain, w…
Active Few-Shot Fine-Tuning
Jonas Hübotter, Bhavya Sukhija, Lenart Treven +2
We study the question: How can we select the right data for fine-tuning to a specific task? We call this data selection problem active fine-tuning and show that it is an instance o…
Efficient Exploration in Continuous-time Model-based Reinforcement Learning
Lenart Treven, Jonas Hübotter, Bhavya Sukhija +2
Reinforcement learning algorithms typically consider discrete-time dynamics, even though the underlying systems are often continuous in time. In this paper, we introduce a model-ba…
A Cut-Matching Game for Constant-Hop Expanders
Bernhard Haeupler, Jonas Huebotter, Mohsen Ghaffari
This paper extends and generalizes the well-known cut-matching game framework and provides a novel cut-strategy that produces constant-hop expanders. Constant-hop expanders are a s…
Implementation of Algorithms for Right-Sizing Data Centers
Jonas Hübotter
The energy consumption of data centers assumes a significant fraction of the world's overall energy consumption. Most data centers are statically provisioned, leading to a very low…