16 citations · 26 across the 14 of their papers we have counts for
18 papers
Weight Pair Encoding: Inducing a Smaller Grammar in Neural Network Weights
Irene Tallini, Daniele Solombrino, Alberto Cazzaniga +1
We show that neural network weights can be explicilty fintuned to admit a smaller grammar. Weight Pair Encoding (WeightPE) does so by placing a lossy Re-Pair compressor inside a st…
LocUS: Head Selection and Subspace Projection for Targeted Activation Steering
Irene Tallini, Lorenzo Basile, Valentino Maiorca +2
Activation steering is a powerful training-free paradigm for controlling large language models at inference time. However, standard approaches estimate a per-layer steering directi…
Visual Instruction Tuning Aligns Modalities through Abstraction
Luis Palacios, Lorenzo Basile, Diego Doimo +1
Visual instruction tuning effectively adapts a pre-trained Large Language Model (LLM) to process image information alongside text. Yet, it remains unclear how visual features are e…
Preserving Historical Truth: Detecting Historical Revisionism in Large Language Models
Francesco Ortu, Joeun Yook, Punya Syon Pandey +5
Large language models (LLMs) are increasingly used as sources of historical information, motivating the need for scalable audits on contested events and politically charged narrati…
Are LLMs Good Safety Agents or a Propaganda Engine?
Neemesh Yadav, Francesco Ortu, Jiarui Liu +5
Large Language Models (LLMs) are trained to refuse to respond to harmful content. However, systematic analyses of whether this behavior is truly a reflection of its safety policies…
Head Pursuit: Probing Attention Specialization in Multimodal Transformers
Lorenzo Basile, Valentino Maiorca, Diego Doimo +2
Language and vision-language models have shown impressive performance across a wide range of tasks, but their internal mechanisms remain only partly understood. In this work, we st…
When Seeing Overrides Knowing: Disentangling Knowledge Conflicts in Vision-Language Models
Francesco Ortu, Zhijing Jin, Diego Doimo +1
Vision-language models (VLMs) increasingly combine visual and textual information to perform complex tasks. However, conflicts between their internal knowledge and external visual…
The Intrinsic Dimension of Prompts in Internal Representations of Large Language Models
Karthik Viswanathan, Yuri Gardinazzi, Giada Panerai +2
We study the geometry of token representations at the prompt level in large language models through the lens of intrinsic dimension. Viewing transformers as mean-field particle sys…
The Narrow Gate: Localized Image-Text Communication in Native Multimodal Models
Alessandro Pietro Serra, Francesco Ortu, Emanuele Panizon +5
Recent advances in multimodal training have significantly improved the integration of image understanding and generation within a unified model. This study investigates how vision-…
Persistent Topological Features in Large Language Models
Yuri Gardinazzi, Karthik Viswanathan, Giada Panerai +3
Understanding the decision-making processes of large language models is critical given their widespread applications. To achieve this, we aim to connect a formal mathematical frame…
The representation landscape of few-shot learning and fine-tuning in large language models
Diego Doimo, Alessandro Serra, Alessio Ansuini +1
In-context learning (ICL) and supervised fine-tuning (SFT) are two common strategies for improving the performance of modern large language models (LLMs) on specific tasks. Despite…
Competition of Mechanisms: Tracing How Language Models Handle Facts and Counterfactuals
Francesco Ortu, Zhijing Jin, Diego Doimo +3
Interpretability research aims to bridge the gap between empirical success and our scientific understanding of the inner workings of large language models (LLMs). However, most exi…
Emergent representations in networks trained with the Forward-Forward algorithm
Niccolò Tosato, Lorenzo Basile, Emanuele Ballarin +3
The Backpropagation algorithm has often been criticised for its lack of biological realism. In an attempt to find a more biologically plausible alternative, the recently introduced…
The geometry of hidden representations of large transformer models
Lucrezia Valeriani, Diego Doimo, Francesca Cuturello +3
Large transformers are powerful architectures used for self-supervised data analysis across various data types, including protein sequences, images, and text. In these models, the…
What is the probability that a random symmetric tensor is close to rank-one?
Alberto Cazzaniga, Antonio Lerario, Andrea Rosana
We address the general problem of estimating the probability that a real symmetric tensor is close to rank-one tensors. Using Weyl's tube formula, we turn this question into a diff…
Higher rank motivic Donaldson-Thomas invariants of via wall-crossing, and asymptotics
Alberto Cazzaniga, Dimbinaina Ralaivaosaona, Andrea T. Ricolfi
We compute, via motivic wall-crossing, the generating function of virtual motives of the Quot scheme of points on , generalising to higher rank a result of Behrend, B…
Framed motivic Donaldson-Thomas invariants of small crepant resolutions
Alberto Cazzaniga, Andrea T. Ricolfi
For an arbitrary integer , we compute -framed motivic PT and DT invariants of small crepant resolutions of toric Calabi-Yau -folds, establishing a "higher rank" vers…
Motivic Donaldson--Thomas invariants of some quantized threefolds
Alberto Cazzaniga, Andrew Morrison, Brent Pym +1
This paper is motivated by the question of how motivic Donaldson--Thomas invariants behave in families. We compute the invariants for some simple families of noncommutative Calabi-…