Stefano Soatto

Samueli Institute — United States

Computer Science · Physical Sciences

69h-index22.0kcitations629works1.92yr avg
OpenAlex↗ORCID↗Google Scholar↗Lab page↗

Accepting Students?

No reports yet. Know if this professor is taking students?

Research Topics

Advanced Vision and Imaging(186), Advanced Image and Video Retrieval Techniques(95), Robotics and Sensor-Based Localization(87), Domain Adaptation and Few-Shot Learning(84), Advanced Neural Network Applications(70)

Publications629 total

Gated KalmaNet: A Fading Memory Layer Through Test-Time Ridge Regression
arXiv (Cornell University)·2025Open Access
Experience-Guided Adaptation of Inference-Time Reasoning Strategies
arXiv (Cornell University)·2025Open Access
MURPHY: Multi-Turn GRPO for Self Correcting Code Generation
ArXiv.org·2025Open Access
Learning to Focus: Focal Attention for Selective and Scalable Transformers
arXiv (Cornell University)·2025Open Access
e1: Learning Adaptive Control of Reasoning Effort
arXiv (Cornell University)·2025Open Access
AI Agents as Universal Task Solvers
ArXiv.org·2025Open Access
Test-Time Defense Against Adversarial Attacks via Stochastic Resonance of Latent Ensembles
arXiv (Cornell University)·2025Open Access
LATTS: Locally Adaptive Test-Time Scaling
arXiv (Cornell University)·2025Open Access
Maximally-Informative Retrieval for State Space Model Generation
arXiv (Cornell University)·2025Open Access
Scaling up Image Segmentation across Data and Tasks
2025
STree: Speculative Tree Decoding for Hybrid State-Space Models
arXiv (Cornell University)·2025Open Access
Robust Planning for Autonomous Driving via Mixed Adversarial Diffusion Predictions
2025
Robust Planning for Autonomous Driving via Mixed Adversarial Diffusion Predictions
arXiv (Cornell University)·2025Open Access
PICASO: Permutation-Invariant Context Composition with State Space Models
arXiv (Cornell University)·2025Open Access
Descriminative-Generative Custom Tokens for Vision-Language Models
arXiv (Cornell University)·2025Open Access
Expansion Span: Combining Fading Memory and Retrieval in Hybrid State Space Models
arXiv (Cornell University)·2024Open Access
Efficient Scaling of Diffusion Transformers for Text-to-Image Generation
arXiv (Cornell University)·2024Open Access
Heat Death of Generative Models in Closed-Loop Learning
2024· 2 cited
On the Viability of Monocular Depth Pre-training for Semantic Segmentation
Lecture notes in computer science·2024· 1 cited
Diffusion Soup: Model Merging for Text-to-Image Diffusion Models
Lecture notes in computer science·2024· 2 cited
page 1 of 32Next →

Frequent Co-authors

Anthony Yezzi(6), Alessandro Chiuso(5), S. Shankar Sastry(5), Yanchao Yang(4), Gianfranco Doretto(4), Yi Ma(4), Daniel Cremers(4), Paolo Favaro(4), Hailin Jin(4), Renè Vidal(4), Andrea Vedaldi(3), Michalis Raptis(3), Alex Wong(3), Brian Fulkerson(2), Stanley Osher(2), Pratik Chaudhari(2), Anna Choromanska(2), Yann LeCun(2), Carlo Baldassi(2), Christian Borgs(2)