Integrated Calendar

Uncertainty quantification and visualization is crucial for the deployment of image restoration models in safety-critical domains, like biological and medical imaging. To date, methods for visualizing uncertainty have mainly focused on per-pixel estimates, which provide limited information. Theoretically, more natural visualizations of uncertainty could be obtained from a principal component analysis (PCA) or from some clustering of the posterior distribution. However, such approaches would require generating numerous samples from the posterior distribution as a first step, which is computationally impractical with today’s SOTA (diffusion-based) posterior samplers. In this talk I will present methods that can output a hierarchical clustering (a tree) or the principal components (PCs) of the posterior in a single forward pass of a neural network. Our methods are both more accurate and orders of magnitude faster than the naïve approach of applying clustering or PCA to posterior samples generated by a conditional generative model. I will illustrate the effectiveness of our methods on multiple inverse problems in imaging, including denoising, inpainting, super-resolution, colorization, and biological image-to-image translation.
The talk will cover joint works with Elias Nehme, Omer Yair, Hila Manor and Rotem Mulayoff.

Transmissible RNA has emerged as a means of communication between organisms, both within and across different kingdoms of life. Donor organisms transmit long base-paired RNA, tRNA-fragments, and other small RNAs to elicit RNAi responses in recipient individuals, affecting their gene expression and phenotypes. Honeybees offer a unique opportunity to study RNA transmission since they possess a transmissible RNA pathway through which they share RNAs between individuals and across generations via the secretion and ingestion of worker- and royal jelly. We hypothesised that members of the gut microbiome exploit the same pathway and transmit RNA to their honeybee host.
We show that RNA originating from a gut-restricted bacterium, Snodgrassella alvi (S. alvi), can be detected in worker- and royal jellies. Endogenous S. alvi RNAs are present also in systemic larval tissues in the absence of bacterial genomic DNA, indicating jelly-mediated microbiome RNA uptake and systemic spread within recipient larvae. Characterisation of transmissible S. alvi RNA reveals enrichment of specific rRNA and tRNA fragments in systemic larval tissues. The transmitted RNA fragments could potentially be involved in RNAi and have the capacity to target honeybee pathogens, such as Nosema and viruses. An expanded transmissible RNA pathway and its potential cooperative roles in honeybee- microbiome interactions will be discussed.

The field of crystal growth and design has been researched thoroughly, specifically the ability to form crystals with tunable dimensions, morphologies, and functional properties. Notably, various crystallographic defects have been found to enhance material properties. For instance, atomic doping alters electrical properties, screw dislocations facilitate spiral crystal growth, while dislocation outcrops and vacancies enhance catalytic activity and strengthen materials. In this talk, I will show how such crystal defects can be utilized to fine-tune a range of physical properties in crystals and act as templates for their growth. I will also highlight examples of crystals formed in nature that serve as a source of inspiration for the design of novel bio-inspired materials with enhanced functional properties.

Agreement testing (aka direct product testing), checks if consistent local information reveals global structure. Beyond its theoretical connections to probabilistic checkable proofs (PCPs), constructing agreement testers is a fundamental combinatorial question that has exciting applications in coding theory and hardness amplification.

In recent work we construct new derandomized `1%-regime' agreement tests. Derandomization of these tests is an important stepping stone towards derandomizing many PCPs, such as the parallel repetition PCP.

We will define agreement tests and give some background on their importance. Then we will see a surprising connection between agreement testing to a problem in algebraic topology. Finally, we will discuss how strong group theoretic tools solve this problem and lead to our construction.

Based on joint work with Irit Dinur and Alex Lubotzky.

Over the last decade, several exciting directions have been initiated by laser-driven plasmas,
e.g., compact particle accelerators, inertial fusion and laboratory astrophysics. The first has
known rapid progress, in terms of current, energy, stability; fusion has gone through a historic
step, with the news of ignition being achieved at NIF in 2022; and laboratory astrophysics has
known also spectacular developments, demonstrating the possibility to perform fully scalable
experiments relevant to various objects such as forming stars and supernovae. A particularly
interesting aspect is that all these fields are strongly synergistic, i.e., that advances in one can
push the others as well. I will present examples of such synergies, through recent results
we have obtained in all these domains, and in particular how ultra-bright neutron beams
can be developed using latest generation multi-PW lasers [1,2]. These could open interesting
perspectives in terms of cargo inspection, but also for fusion plasma measurements.
I will also show how fusion can benefit from external magnetization [3]. Finally, I will discuss
advances in laboratory astrophysics, particularly the first-stage acceleration of ions leading to
cosmic rays [4,5], understanding the universal nature of collimated outflows in the Universe [6],
and probing the intricacy of 3D magnetic reconnection [7]
[1] High-flux neutron generation by laser-accelerated ions from single-and double-layer targets, V Horný et al.,
Scientific Reports 12 (1), 19767, 2022
[2] Numerical investigation of spallation neutrons generated from petawatt-scale laser-driven proton beams,
B Martinez et al., Matter and Radiation at Extremes 7 (2), 024401, 2022
[3] Dynamics of nanosecond laser pulse propagation and of associated instabilities in a magnetized underdense
plasma, W. Yao et al., https://doi.org/10.48550/arXiv.2211.06036
[4] Laboratory evidence for proton energization by collisionless shock surfing, W Yao et al.,
Nature Physics 17 (10), 1177-1182, 2021
[5] Enhancement of the Nonresonant Streaming Instability by Particle Collisions, A Marret et al.,
Physical Review Letters 128 (11), 115101, 2022
[6] Laboratory disruption of scaled astrophysical outflows by a misaligned magnetic field, G Revet et al.,
Nature communications 12 (1), 762, 2021
[7] Laboratory evidence of magnetic reconnection hampered in obliquely interacting flux tubes, S Bolaños et al.,
Nature Communications 13 (1), 6426, 2022

LLMs offer users intuitive interfaces for interacting with textual information. The integration of vision into LLMs through VLMs has enabled these models to "see" and reason over visual content. However, these VLMs possess generic knowledge, lacking a personal touch. This raises an intriguing question: can we equip these models with the ability to comprehend and utilize user-specific concepts, tailored specifically to you? Can we ask the model questions about you, such as what you are wearing or what your friend is doing in the image? In this talk, we will explore how we can personalize VLMs to each user, offering more meaningful interactions that better reflect individual experiences and relationships.

Bio:

Yuval is a PhD student at Tel Aviv University under the supervision of Prof. Daniel Cohen-Or. His research centers around leveraging generative models to give users greater control and creative freedom when interacting with visual content. Currently interning at Snap Research under Kfir Aberman, Yuval is also exploring new approaches for personalizing generative models.