Leonardo Ranaldi
YOU?
Author Swipe
View article: Survey on the Role of Mechanistic Interpretability in Generative AI
Survey on the Role of Mechanistic Interpretability in Generative AI Open
The rapid advancement of artificial intelligence (AI) and machine learning has revolutionised how systems process information, make decisions, and adapt to dynamic environments. AI-driven approaches have significantly enhanced efficiency a…
View article: Dissecting Clinical Reasoning in Language Models:A Comparative Study of Prompts and Model Adaptation Strategies
Dissecting Clinical Reasoning in Language Models:A Comparative Study of Prompts and Model Adaptation Strategies Open
Recent works on large language models (LLMs) have demonstrated the impact of prompting strategies and fine-tuning techniques on their reasoning capabilities. Yet, their effectiveness on clinical natural language inference (NLI) remains und…
View article: Animate, or Inanimate, That Is the Question for Large Language Models
Animate, or Inanimate, That Is the Question for Large Language Models Open
The cognitive core of human beings is closely connected to the concept of animacy, which significantly influences their memory, vision, and complex language comprehension. While animacy is reflected in language through subtle constraints o…
View article: Protoknowledge Shapes Behaviour of LLMs in Downstream Tasks: Memorization and Generalization with Knowledge Graphs
Protoknowledge Shapes Behaviour of LLMs in Downstream Tasks: Memorization and Generalization with Knowledge Graphs Open
We introduce the concept of protoknowledge to formalize and measure how sequences of tokens encoding Knowledge Graphs are internalized during pretraining and utilized at inference time by Large Language Models (LLMs). Indeed, LLMs have dem…
View article: Improving Multilingual Retrieval-Augmented Language Models through Dialectic Reasoning Argumentations
Improving Multilingual Retrieval-Augmented Language Models through Dialectic Reasoning Argumentations Open
Retrieval-augmented generation (RAG) is key to enhancing large language models (LLMs) to systematically access richer factual knowledge. Yet, using RAG brings intrinsic challenges, as LLMs must deal with potentially conflicting knowledge, …
View article: Multilingual Retrieval-Augmented Generation for Knowledge-Intensive Task
Multilingual Retrieval-Augmented Generation for Knowledge-Intensive Task Open
Retrieval-augmented generation (RAG) has become a cornerstone of contemporary NLP, enhancing large language models (LLMs) by allowing them to access richer factual contexts through in-context retrieval. While effective in monolingual setti…
View article: MeMo: Towards Language Models with Associative Memory Mechanisms
MeMo: Towards Language Models with Associative Memory Mechanisms Open
Memorization is a fundamental ability of Transformer-based Large Language Models, achieved through learning. In this paper, we propose a paradigm shift by designing an architecture to memorize text directly, bearing in mind the principle t…
View article: Improving Chain-of-Thought Reasoning via Quasi-Symbolic Abstractions
Improving Chain-of-Thought Reasoning via Quasi-Symbolic Abstractions Open
Chain-of-Though (CoT) represents a common strategy for reasoning in Large Language Models (LLMs) by decomposing complex tasks into intermediate inference steps. However, explanations generated via CoT are susceptible to content biases that…
View article: Eliciting Critical Reasoning in Retrieval-Augmented Language Models via Contrastive Explanations
Eliciting Critical Reasoning in Retrieval-Augmented Language Models via Contrastive Explanations Open
Retrieval-augmented generation (RAG) has emerged as a critical mechanism in contemporary NLP to support Large Language Models(LLMs) in systematically accessing richer factual context. However, the integration of RAG mechanisms brings its i…
View article: Animate, or Inanimate, That is the Question for Large Language Models
Animate, or Inanimate, That is the Question for Large Language Models Open
The cognitive essence of humans is deeply intertwined with the concept of animacy, which plays an essential role in shaping their memory, vision, and multi-layered language understanding. Although animacy appears in language via nuanced co…
View article: Self-Refine Instruction-Tuning for Aligning Reasoning in Language Models
Self-Refine Instruction-Tuning for Aligning Reasoning in Language Models Open
The alignments of reasoning abilities between smaller and larger Language Models are largely conducted via Supervised Fine-Tuning (SFT) using demonstrations generated from robust Large Language Models (LLMs). Although these approaches deli…
View article: Investigating the Impact of Data Contamination of Large Language Models in Text-to-SQL Translation
Investigating the Impact of Data Contamination of Large Language Models in Text-to-SQL Translation Open
Understanding textual description to generate code seems to be an achieved capability of instruction-following Large Language Models (LLMs) in zero-shot scenario. However, there is a severe possibility that this translation ability may be …
View article: Self-Refine Instruction-Tuning for Aligning Reasoning in Language Models
Self-Refine Instruction-Tuning for Aligning Reasoning in Language Models Open
The alignment of reasoning abilities between smaller and larger Language Models are largely conducted via supervised fine-tuning using demonstrations generated from robust Large Language Models (LLMs). Although these approaches deliver mor…
View article: Empowering Multi-step Reasoning across Languages via Program-Aided Language Models
Empowering Multi-step Reasoning across Languages via Program-Aided Language Models Open
In-context learning methods are commonly employed as inference strategies, where Large Language Models (LLMs) are elicited to solve a task by leveraging provided demonstrations without requiring parameter updates. Among these approaches ar…
View article: Empowering cross-lingual abilities of instruction-tuned large language models by translation-following demonstrations
Empowering cross-lingual abilities of instruction-tuned large language models by translation-following demonstrations Open
The language ability of Large Language Models (LLMs) is often unbalanced towards English because of the imbalance in the distribution of the pre-training data. This disparity is demanded in further fine-tuning and affecting the cross-lingu…
View article: Investigating the Impact of Data Contamination of Large Language Models in Text-to-SQL translation
Investigating the Impact of Data Contamination of Large Language Models in Text-to-SQL translation Open
Understanding textual description to generate code seems to be an achieved capability of instruction-following Large Language Models (LLMs) in zero-shot scenario. However, there is a severe possibility that this translation ability may be …
View article: When Large Language Models contradict humans? Large Language Models' Sycophantic Behaviour
When Large Language Models contradict humans? Large Language Models' Sycophantic Behaviour Open
Large Language Models have been demonstrating broadly satisfactory generative abilities for users, which seems to be due to the intensive use of human feedback that refines responses. Nevertheless, suggestibility inherited via human feedba…
View article: Empowering Multi-step Reasoning across Languages via Tree-of-Thoughts
Empowering Multi-step Reasoning across Languages via Tree-of-Thoughts Open
Reasoning methods, best exemplified by the well-known Chain-of-Thought (CoT), empower the reasoning abilities of Large Language Models (LLMs) by eliciting them to solve complex tasks in a step-by-step manner. Although they are achieving si…
View article: The Dark Side of the Language: Syntax-Based Neural Networks Rivaling Transformers in Definitely Unseen Sentences
The Dark Side of the Language: Syntax-Based Neural Networks Rivaling Transformers in Definitely Unseen Sentences Open
Syntax-based methods have been largely used as key components of Natural Language Processing systems for solving a variety of tasks. Yet, pre-trained Transformers are challenging all these pre-existing methods and even humans in nearly all…
View article: HANS, are you clever? Clever Hans Effect Analysis of Neural Systems
HANS, are you clever? Clever Hans Effect Analysis of Neural Systems Open
Instruction-tuned Large Language Models (It-LLMs) have been exhibiting outstanding abilities to reason around cognitive states, intentions, and reactions of all people involved, letting humans guide and comprehend day-to-day social interac…