Papers by Alberto Testoni
ACT-Thor: A Controlled Benchmark for Embodied Action Understanding in Simulated Environments (2022.coling-1)
Copied to clipboard
| Challenge: | embodied AI tasks require a strong understanding of verbs and their corresponding actions. |
| Approach: | They propose a controlled benchmark for embodied action understanding using a simulated environment and a visual feature extractor. |
| Outcome: | The proposed benchmark achieves 81.4% accuracy and high inter-annotator agreement . the proposed model falls behind human models in a zero-shot scenario . |
LLMs instead of Human Judges? A Large Scale Empirical Study across 20 NLP Evaluation Tasks (2025.acl-short)
Copied to clipboard
Anna Bavaresco, Raffaella Bernardi, Leonardo Bertolazzi, Desmond Elliott, Raquel Fernández, Albert Gatt, Esam Ghaleb, Mario Giulianelli, Michael Hanna, Alexander Koller, Andre Martins, Philipp Mondorf, Vera Neplenbroek, Sandro Pezzelle, Barbara Plank, David Schlangen, Alessandro Suglia, Aditya K Surikuchi, Ece Takmaz, Alberto Testoni
| Challenge: | Existing evaluations of NLP models with LLMs are based on human judgments . however, there are concerns about their validity and reproducibility in proprietary models . |
| Approach: | They evaluate 11 current LLMs for their ability to replicate annotations. they show substantial variance across models and datasets. |
| Outcome: | The proposed model can replicate human annotations on 20 NLP datasets and show substantial variance across models and datasets. |
Grounded Textual Entailment (C18-1)
Copied to clipboard
Hoa Trong Vu, Claudio Greco, Aliia Erofeeva, Somayeh Jafaritazehjan, Guido Linders, Marc Tanti, Alberto Testoni, Raffaella Bernardi, Albert Gatt
| Challenge: | Existing models for entailment analysis are not performing well in visual information-based models. |
| Approach: | They propose to use a visual representation of the Textual Entailment task to compare visual-grounded models with a multimodal version of the SNLI dataset. |
| Outcome: | The proposed model performs better when there is an image of the “world” or “situation” . |
Mind the Gap: Benchmarking LLM Uncertainty and Calibration with Specialty-Aware Clinical QA and Reasoning-Based Behavioural Features (2026.eacl-long)
Copied to clipboard
| Challenge: | Reliable uncertainty quantification (UQ) is essential when employing large language models in high-risk domains such as clinical question answering (QA). |
| Approach: | They evaluate uncertainty estimation methods for clinical question answering using eleven clinical specialties and six question types. |
| Outcome: | The proposed method is based on behavioral features derived from reasoning-oriented models and examines conformal prediction as a complementary set-based approach. |
Asking the Right Question at the Right Time: Human and Model Uncertainty Guidance to Ask Clarification Questions (2024.eacl-long)
Copied to clipboard
| Challenge: | Using model uncertainty as supervision for deciding when to ask may not be the most effective way to resolve model uncertainty. |
| Approach: | They propose to generate clarification questions based on model uncertainty estimation and compare it to several alternatives to generate questions . |
| Outcome: | The proposed approach improves the model uncertainty of a collaborative dialogue task and shows that it is more effective than other alternatives. |
RAcQUEt: Unveiling the Dangers of Overlooked Referential Ambiguity in Visual LLMs (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing language models that address ambiguity are limited in their ability to address it . ambiguities are an inherent feature of human language, according to research . |
| Approach: | They propose a dataset targeting referential ambiguity in image-based question answering . they find that current language models lack robust strategies to deal with ambiguities . |
| Outcome: | The proposed dataset shows that state-of-the-art models fail to address ambiguity . ambiguities are an inherent feature of human language, according to research . |
Learning to Ask Informative Questions: Enhancing LLMs with Preference Optimization and Expected Information Gain (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Large language models (LLMs) often perform poorly in generating informative questions, as measured by expected information gain (EIG). |
| Approach: | They propose to use a large language model to enhance the informativeness of LLM-generated questions in 20-question game dialogues by applying a Direct Preference Optimization algorithm to generate low-EIG and high-EI questions. |
| Outcome: | The proposed method produces more effective questions even in domains different from those used to train the DPO model. |
From Tools to Teammates: Evaluating LLMs in Multi-Session Coding Interactions (2025.acl-long)
Copied to clipboard
Nathanaël Carraz Rakotonirina, Mohammed Hamdy, Jon Ander Campos, Lucas Weber, Alberto Testoni, Marzieh Fadaee, Sandro Pezzelle, Marco Del Tredici
| Challenge: | Large Language Models excel at solving individual problems in isolation, but are they able to effectively collaborate over long-term interactions? |
| Approach: | They propose to use a multi-session dataset to test LLMs' ability to track and execute simple coding instructions amid irrelevant information, simulating a realistic setting. |
| Outcome: | The proposed model performs poorly when instructions are spread across sessions, suggesting that they are not able to integrate information over long interactions. |
Naming, Describing, and Quantifying Visual Objects in Humans and LLMs (2024.acl-short)
Copied to clipboard
| Challenge: | Recent work has highlighted that speakers display a wide range of variability when asked to utter sentences, resulting in inter-speaker variability but also variability over time for the same speaker. |
| Approach: | They evaluate Vision & Language Large Language Models (VLLMs) on three categories where humans show great subjective variability concerning the distribution over plausible labels. |
| Outcome: | The proposed models can mimic human distributions over plausible labels, but fail to assign quantifiers, a task that requires more accurate, high-level reasoning. |
Playpen: An Environment for Exploring Learning From Dialogue Game Feedback (2025.emnlp-main)
Copied to clipboard
Nicola Horst, Davide Mazzaccara, Antonia Schmidt, Michael Sullivan, Filippo Momentè, Luca Franceschetti, Philipp Sadler, Sherzod Hakimov, Alberto Testoni, Raffaella Bernardi, Raquel Fernández, Alexander Koller, Oliver Lemon, David Schlangen, Mario Giulianelli, Alessandro Suglia
| Challenge: | In this paper, we investigate whether Dialogue Games—goal-directed and rule-governed activities driven predominantly by verbal actions—can also serve as a source of feedback signals for learning. |
| Approach: | They introduce Playpen, an environment for off- and online learning through Dialogue Game self-play, and investigate a representative set of post-training methods: supervised fine-tuning, direct alignment and reinforcement learning with Group Relative Policy Optimization. |
| Outcome: | The proposed model improves performance on unseen instances, but negatively impacts other skills, while interactive learning shows balanced improvements without loss of skills. |
Calibrated? Not for Everyone: How Sexual Orientation and Religious Markers Distort LLM Accuracy and Confidence in Medical QA (2026.acl-short)
Copied to clipboard
| Challenge: | a growing body of research suggests that social descriptors can influence LLM-generated clinical recommendations. |
| Approach: | They examine whether social descriptors of a patient distort uncertainty signals and model accuracy. |
| Outcome: | The presence of social identity cues affects the reliability of confidence signals, the authors show . incorporating sociodemographic attributes alters outputs in clinical trial matching and QA . |
Looking for Confirmations: An Effective and Human-Like Visual Dialogue Strategy (2021.emnlp-main)
Copied to clipboard
| Challenge: | State-of-the-art systems generate questions that sound unnatural to humans and are grammatically correct. |
| Approach: | They propose to use beam search re-ranking to generate a model that guides an effective goal-oriented strategy by asking questions that confirm the model’s conjecture about the referent. |
| Outcome: | The proposed model is more natural and effective than beam search decoding without re-ranking on the GuessWhat?! game. |
The Interplay of Task Success and Dialogue Quality: An in-depth Evaluation in Task-Oriented Visual Dialogues (2021.eacl-main)
Copied to clipboard
| Challenge: | chit-chat and task-oriented dialogue models are evaluated on their task success metric, but the best model is usually chosen based on task success. |
| Approach: | They compare models playing different games to find out which one is best . they find that this discrepancy is model- and task-agnostic . |
| Outcome: | The proposed model can generate utterances that are indistinguishable from human dialogues by learning to ground, encode, and decode words that do not occur frequently in the training set. |
“I’ve Seen Things You People Wouldn’t Believe”: Hallucinating Entities in GuessWhat?! (2021.acl-srw)
Copied to clipboard
| Challenge: | a problem with natural language generation systems is the generation of tokens that are unrelated to the source input. |
| Approach: | They propose two new models to play the GuessWhat?! referential game . they propose to adapt the best visual processing models available to mitigate this issue . |
| Outcome: | The proposed models generate few hallucinations compared to other models available in the literature. |