1st Year joint workshop of FAIR and ELLIS
FBK - Sala Stringa, December, 01, 2023
KEYNOTE
Cross-modal generation of multimodal content
Nicu Sebe - University of Trento
Abstract:
Video generation consists of generating a video sequence so that an object in a source image is animated according to some external information (a conditioning label, a driving video, a piece of text). In this talk I will present some of our recent achievements addressing generating videos without using any annotation or prior information about the specific object to animate. Once trained on a set of videos depicting objects of the same category (e.g. faces, human bodies), our method can be applied to any object of this class. Based on this, I will further present a framework to train game-engine-like neural models, solely from monocular annotated videos. Similarly to a game engine, it models the logic of the game and the underlying rules of physics, to make it possible for a user to play the game by specifying both high- and low-level action sequences. This requires learning the game's AI, encapsulated by the animation model, to navigate the scene using high-level constraints, play against an adversary, devise the strategy to win a point. I will also highlight the limitation and propose some ideas for future research.
Program
click here to register (closed)
REGISTRATION 8:30 - 9:00
MORNING (9:00 - 12:00)
9.00 - 9.15 Paolo Traverso & Paolo Giorgini: Welcome
Chair Alessandro Sperduti
9.15 - 10.15 Keynote: Nicu Sebe: Cross-modal generation of multimodal content
Chair: Bruno Lepri
10.30 - 11.00 WP2.3: Francesca Meneghello: Runtime Integration of Machine Learning and Simulation for Business Processes. Francesca Meneghello, Chiara Di Francescomarino and Chiara Ghidini, 5th International Conference on Process Mining (ICPM), 2023.
11.00 - 11:30 WP2.2 Tommaso Campari: Exploiting Proximity-Aware Tasks for Embodied Social Navigation. Enrico Cancelli*, Tommaso Campari*, Luciano Serafini, Angel X. Chang, Lamberto Ballan - International conference on Computer Vision - ICCV23, 2023.
11:30 - 12:00 WP2.5 Beatrice Savoldi: Hi Guys or Hi Folks? Benchmarking Gender-Neutral Machine Translation with the GeNTE Corpus. Andrea Piergentili, Beatrice Savoldi, Dennis Fucci, Matteo Negri, Luisa Bentivogli. The 2023 Conference on Empirical Methods in Natural Language Processing - EMNLP 2023
LUNCH WITH POSTERS (12:00 - 14:00)
Mousavi. M. S., Tanaka, S., Roccabruna G., Yoshino K., Nakamura S. and Riccardi G., “ What’s New? Identifying the Unfolding of New Events in a Narrative ” ACL, Proc. 5th Workshop of Narrative Understanding, 2023.
Roccabruna G., Mousavi. M. S. and Riccardi G., “ Understanding Emotion Valence is a Joint Deep Learning Task” ACL, Proc. 13th Workshop on Computational Approaches to Subjectivity, Sentiment & Social Media Analysis 2023.
The Unreasonable Effectiveness of Large Language-Vision Models for Source-Free Video Domain Adaptation Giacomo Zara, Alessandro Conti, Subhankar Roy, Stéphane Lathuilière, Paolo Rota, Elisa Ricci, ICCV 2023
Object-aware Gaze Target Detection Francesco Tonini, Nicola Dall'Asen, Cigdem Beyan, Elisa Ricci, ICCV 2023
Learning to Act for Perceiving in Partially Unknown Environments L Lamanna, M Faridghasemnia, A Gerevini, A Saetti, A Saffiotti, L Serafini, and P. Traverso, IJCAI 2023
Deep Symbolic Learning: Discovering Symbols and Rules from Perceptions A Daniele, T Campari, S Malhotra, L Serafini IJCAI 2023
On CNF Conversion for Disjoint SAT Enumeration. Presentato da Gabriele Masina -- DISI, UNITN
A Unified Framework for Probabilistic Verification of AI Systems via Weighted Model Integration - Paolo Morettin -- DISI, UNITN
Aerial robotics to support human workers in harvesting wild fruits in complex environments - Fabio Remondino and Pawel Trybala FBK
"Integrating Language Models into Direct Speech Translation: An Inference-Time Solution to Control Gender Inflection". Dennis Fucci, Marco Gaido, Sara Papi, Mauro Cettolo, Matteo Negri, Luisa Bentivogli. EMNLP 2023.arXiv
"Behavioural maps for assited navigation". Placido Falqueto
Ancilotto, Alberto, Francesco Paissan, and Elisabetta Farella. "XiNet: Efficient Neural Networks for tinyML." Proceedings of the IEEE/CVF International Conference on Computer Vision. 2023.
Challenging Large Multimodal Models to Reason in Real Situations” Raffaella Bernardi, Bernardo Magnini, Giovanni Bonetta
Process Knowledge Extraction and Knowledge Graph Construction Through Prompting: A Quantitative Analysis. Patrizio Bellan, Mauro Dragoni and Chiara GhidiniSAC 2024
Learning the optimal policy from temporal process data: a white box approach. Massimiliano Ronzani
Trajectory test-train overlap in next-location prediction datasets. M Luca, L Pappalardo, B Lepri, G Barlacchi ECML-PKDD, Journal Track Machine Learning
The benefits of sharing: a cloud-aided performance-oriented framework to learn optimal feedback policies. L. Ferrarotti, V. Breschi, A. Bemporad. In Proc. of Learning for Dynamics and Control
AFTERNOON (14:00-17:00
Chair: Paolo Giorgini
14.00 - 14.30 WP2.1 Gabriele Masina: On CNF Conversion for Disjoint SAT Enumeration. Gabriele Masina, Giuseppe Spallitta, Roberto Sebastiani. 26th International Conference on Theory and Applications of Satisfiability Testing -- SAT'23. 2023.
14.30 - 13.00 WP2.6 Massimo Zancanaro: AI agents in Virtual Reality as an educational tool: assessing verbal and behavioral responses in virtual and real conversations. Ersilia Vallefuoco, Giovanna Paola Varni, Massimo Zancanaro. In progress 2023
15.00 - 15.30 WP2.4 Massimiliano Mancini: Vocabulary-free Image Classification. Alessandro Conti, Enrico Fini, Massimiliano Mancini, Palo Rota, Yiming Wang, Elisa Ricci. NeurIPS 2023
15.30 - 15.45 Coffee Break
Chair: Luciano Serafini
15.45 - 16.15 WP2.9 Emanuele Marconato: Not All Neuro-Symbolic Concepts Are Created Equal: Analysis and Mitigation of Reasoning Shortcuts. Emanuele Marconato, Stefano Teso, Antonio Vergari, Andrea Passerini. NeurIPS 2023.
16.15 - 16.45 WP2.7 Luigi Palopoli: Will we have robot doctors in the future? A pragmatic perspectives toward robotised ultrasound diagnosis. Submitted to ICRA 2023.
16.45 - 17.00 Paolo Travreso and Paolo Giorgini Conclusions
Some pictures:
