4 citations · 7 across the 11 of their papers we have counts for
13 papers
How Depth Growth and Looping Shape In-Context Learning in Language Models
Ferdinand Kapl, Kaitlin Maile, Stefan Bauer +1
Looped and depth-grown language models reuse computations across depth and have attracted interest for their reasoning capabilities. Building on prior suggestive evidence of improv…
Growing to See: Efficient Pretraining of Self-Supervised Vision Transformers
Amir Mohammad Karimi Mamaghan, Ferdinand Kapl, Andrea Dittadi +4
Progressive depth growth builds large networks by starting with shallow models and inserting layers during training. Recent advances in large language models have renewed interest…
Evolution With Purpose: Hierarchy-Informed Optimization of Whole-Brain Models
Hormoz Shahrzad, Niharika Gajawelli, Kaitlin Maile +2
Evolutionary search is well suited for large-scale biophysical brain modeling, where many parameters with nonlinear interactions and no tractable gradients need to be optimized. St…
From Growing to Looping: A Unified View of Iterative Computation in LLMs
Ferdinand Kapl, Emmanouil Angelis, Kaitlin Maile +2
Looping, reusing a block of layers across depth, and depth growing, training shallow-to-deep models by duplicating middle layers, have both been linked to stronger reasoning, but t…
From Words to Amino Acids: Does the Curse of Depth Persist?
Aleena Siji, Amir Mohammad Karimi Mamaghan, Ferdinand Kapl +9
Protein language models (PLMs) have become widely adopted as general-purpose models, demonstrating strong performance in protein engineering and de novo design. Like large language…
Emergent temporal abstractions in autoregressive models enable hierarchical reinforcement learning
Seijin Kobayashi, Yanick Schimpf, Maximilian Schlegel +12
Large-scale autoregressive models pretrained on next-token prediction and finetuned with reinforcement learning (RL) have achieved unprecedented success on many problem domains. Du…
Best Practices For Empirical Meta-Algorithmic Research: Guidelines from the COSEAL Research Network
Theresa Eimer, Lennart Schäpermeier, André Biedenkapp +15
Empirical research on meta-algorithmics, such as algorithm selection, configuration, and scheduling, often relies on extensive and thus computationally expensive experiments. With…
Do Depth-Grown Models Overcome the Curse of Depth? An In-Depth Analysis
Ferdinand Kapl, Emmanouil Angelis, Tobias Höppe +4
Gradually growing the depth of Transformers during training can not only reduce training cost but also lead to improved reasoning performance, as shown by MIDAS (Saunshi et al., 20…
MesaNet: Sequence Modeling by Locally Optimal Test-Time Training
Johannes von Oswald, Nino Scherrer, Seijin Kobayashi +14
Sequence modeling is currently dominated by causal transformer architectures that use softmax self-attention. Although widely adopted, transformers require scaling memory and compu…
Composable Function-preserving Expansions for Transformer Architectures
Andrea Gesmundo, Kaitlin Maile
Training state-of-the-art neural networks requires a high cost in terms of compute and time. Model scale is recognized to be a critical factor to achieve and improve the state-of-t…
Equivariance-aware Architectural Optimization of Neural Networks
Kaitlin Maile, Dennis G. Wilson, Patrick Forré
Incorporating equivariance to symmetry groups as a constraint during neural network training can improve performance and generalization for tasks exhibiting those symmetries, but s…
When, where, and how to add new neurons to ANNs
Kaitlin Maile, Emmanuel Rachelson, Hervé Luga +1
Neurogenesis in ANNs is an understudied and difficult problem, even compared to other forms of structural learning like pruning. By decomposing it into triggers and initializations…
DARTS-PRIME: Regularization and Scheduling Improve Constrained Optimization in Differentiable NAS
Kaitlin Maile, Erwan Lecarpentier, Hervé Luga +1
Differentiable Architecture Search (DARTS) is a recent neural architecture search (NAS) method based on a differentiable relaxation. Due to its success, numerous variants analyzing…