Publications


Here is a selection of recent work. For a complete list, check my Google Scholar.

Adaptive LLM-Symbolic Reasoning via Dynamic Logical Solver Composition

EACL 2026

Lei Xu, Pierre Beckmann, Marco Valentino, Andre Freitas

Can Activation Steering Generalize Across Languages? A Study on Syllogistic Reasoning in Language Models

EACL 2026

Gabriele Maraia, Leonardo Ranaldi, Marco Valentino, Fabio Massimo Zanzotto

Mitigating Content Effects on Reasoning in Language Models Through Fine-Grained Activation Steering

AAAI 2026 (Oral)

Marco Valentino, Geonhee Kim, Dhairya Dalal, Zhixue Zhao, André Freitas

Autoformalization in the Wild: Assessing LLMs on Real-World Mathematical Definitions 🏆 Best Resource Paper

EMNLP 2025

Lan Zhang, Marco Valentino, André Freitas

Enhancing Logical Reasoning in Language Models via Symbolically-Guided Monte Carlo Process Supervision

EMNLP 2024

Xingwei Tan, Marco Valentino, Mahmud Elahi Akhter, Maria Liakata, Nikolaos Aletras

Unravelling the Logic: Investigating the Generalisation of Transformers in Numerical Satisfiability Problems

ACL 2025

Tharindu Madusanka, Marco Valentino, Iqra Zahid, Ian Pratt-Hartmann, Riza Batista-Navarro

Faithful and Robust LLM-Driven Theorem Proving for NLI Explanations

ACL 2025

Xin Quan, Marco Valentino, Louise Dennis, André Freitas

Improving Chain-of-Thought Reasoning via Quasi-Symbolic Abstractions

ACL 2025

Leonardo Ranaldi, Marco Valentino, André Freitas

PEIRCE: Unifying Material and Formal Reasoning via LLM-Driven Neuro-Symbolic Refinement

ACL 2025 (Demo Track)

Xin Quan, Marco Valentino, Danilo S Carvalho, Dhairya Dalal, André Freitas

Reasoning Circuits in Language Models: A Mechanistic Interpretation of Syllogistic Inference

ACL 2025 (Findings)

Geonhee Kim, Marco Valentino, André Freitas

Controlling Equational Reasoning in Large Language Models with Prompt Interventions

AAAI 2025 (Oral)

Jordan Meadows, Marco Valentino, André Freitas

Eliciting Critical Reasoning in Retrieval-Augmented Generation via Contrastive Explanations

NAACL 2025

Leonardo Ranaldi, Marco Valentino, André Freitas

On the Nature of Explanation: An Epistemological‑Linguistic Perspective for Explanation‑Based Natural Language Inference

Philosophy & Technology

Marco Valentino, André Freitas

Verification and Refinement of Natural Language Explanations through LLM-Symbolic Theorem Proving 🏆 Outstanding Paper

EMNLP 2024

Xin Quan, Marco Valentino, Louise Dennis, André Freitas

Inference to the Best Explanation in Large Language Models

ACL 2024

Dhairya Dalal, Marco Valentino, André Freitas, Paul Buitelaar

Multi-Operational Mathematical Derivations in Latent Space

NAACL 2024

Marco Valentino, Jordan Meadows, Lan Zhang, André Freitas

Multi-Relational Hyperbolic Word Embeddings from Natural Language Definitions

EACL 2024

Marco Valentino, Danilo S. Carvalho, André Freitas

NLI4CT: Multi-Evidence Natural Language Inference for Clinical Trial Reports

EMNLP 2023

Mael Jullien, Marco Valentino, Hannah Frost, Paul O'Regan, Donal Landers, André Freitas

Awards

EMNLP 2025 - Best Resource Paper Award

"Autoformalization in the Wild: Assessing LLMs on Real-World Mathematical Definitions". Lan Zhang, Marco Valentino, André Freitas.

EMNLP 2024 - Outstanding Paper Award

"Verification and Refinement of Natural Language Explanations through LLM-Symbolic Theorem Proving". Xin Quan, Marco Valentino, Louise Dennis, André Freitas.

SemEval-2023 - Best Task Paper Award Honourable Mention

"SemEval-2023 Task 7: Multi-Evidence Natural Language Inference for Clinical Trial Data". Maël Jullien, Marco Valentino, Hannah Frost, Paul O’regan, Donal Landers, André Freitas.