6 papers
Looped Language Models Improve Compositional Tool Calling
Andrei Cristian Popescu, Haitz Sáez de Ocáriz Borde, Pietro Liò
Looped language models have shown promising results on reasoning benchmarks, yet their potential for agentic tool use remains largely unexplored. We study this question in composit…
Adaptive Depth in Looped Transformers: Diagnosing Learned Halting Gates and Trajectory Readouts
Andrei Cristian Popescu, Haitz Sáez de Ocáriz Borde, Pietro Liò
Looped Transformers increase test-time computation by repeatedly applying a shared recurrent block. Learned halting objectives in looped Transformers typically use a single exit di…
Algebraic Priors for Approximately Equivariant Networks
Riccardo Ali, Pietro Liò, Jamie Vicary
Equivariant neural networks incorporate symmetries through group actions, embedding them as an inductive bias to improve performance. Existing methods learn an equivariant action o…
The Stepwise Informativeness Assumption: Why are Entropy Dynamics and Reasoning Correlated in LLMs?
Mar Gonzà lez I CatalÃ, Haitz Sáez de Ocáriz Borde, George D. Montañez +1
Recent work uses entropy-based signals at multiple representation levels to study reasoning in large language models, but the field remains largely empirical. A central unresolved…
Entropy-Lens: Uncovering Decision Strategies in LLMs
Riccardo Ali, Francesco Caso, Christopher Irwin +1
In large language models (LLMs), each block operates on the residual stream to map input token sequences to output token distributions. However, most of the interpretability litera…
Assessing the Geographic Generalization and Physical Consistency of Generative Models for Climate Downscaling
Carlo Saccardi, Maximilian Pierzyna, Haitz Sáez de Ocáriz Borde +6
Kilometer-scale weather data is crucial for real-world applications but remains computationally intensive to produce using traditional weather simulations. An emerging solution is…