Dual encoders are now the dominant architecture for dense retrieval. Yet, we have little understanding of how they represent text, and why this leads to good performance. Here we shed light on this question via distributions over the vocabulary. We propose to interpret the vector representations produced by dual encoders by projecting them into the model's vocabulary space. We show that the resulting distributions over vocabulary tokens are intuitive and contain rich semantic information. We find that this view can explain some of the failure cases of dense retrievers. For example, the inability of models to handle tail entities can be explained via a tendency of the token distributions to forget some of the tokens of those entities. We leverage this insight and propose a simple way to enrich query and passage representations with lexical information at inference time, and show that this significantly improves performance compared to the original model in out-of-domain settings.
Joint work with: Ori Ram, Liat Bezalel, Adi Zicher, Yonatan Belinkov and Jonathan Berant.
 

Holds a BSc in computer science and physics, and a PhD in computational neuroscience. After his PhD, he was a postdoctoral fellow at the University of Toronto and a postdoctoral fellow at MIT. His research interests include machine learning, deep learning, graphical models, optimization, machine vision, and natural language processing. His work has received several prizes including five paper awards at NeurIPS, ICML and UAI. In 2019, he received the ERC Consolidator Grant.

Drug discovery and development is a lengthy and expensive process with more than a 90% failure rate. Currently, most methods for assessing preclinical safety of a new drug candidate rely on animal testing, which fails to reliably predict the interaction of the drug with the human body. More than that, legislation in the US and EU aims to stop animals from being used for scientific purposes. Indeed, the FDA recently canceled the mandating of animal testing in the approval process of a new drug. Thus, the development of new methods for the prediction of clinical safety is a global, major unmet need. In this talk, I'll present a BIO-AI prediction platform for clinical safety. The platform integrates state-of-the-art machine learning, with patient-on-a-chip (miniaturized human organs), real-time nano-sensing, and stem-cell genomic diversity technologies. The platform performs millions of AI-guided interactions between genetically diverse patient-on-a-chip and potential drugs. While observing the interactions, the platform generates unique data, highly indicative of drug toxicities. The platform-generated data is used for dynamically training multimodal deep learning to predict drug safety in a multi-task fashion. The platform is currently used by big pharmaceutical companies for the safety assessment of new drug candidates. It will be used in the future to generate recommendations for personal safety and treatment using approved drugs.

Isaac is the Founder and CEO of Quris. Prior to Quris, Isaac founded and led three bio-AI technology companies, each of which led revolutions in medicine, genomics, agriculture and conservation. Isaac is a physician and entrepreneur with a passion for leading interdisciplinary teams of scientists and technologists to tackle impactful challenges in the intersection between machine-learning and life-sciences; and to leverage and commercialize the resulting solutions. One of the companies that he founded was Rosetta Genomics (NASDAQ:ROSG), which analyzed the human genome. Isaac led the team at Rosetta Genomics in the discovery of hundreds of novel genes, more so than all the universities in the world combined, and delivered novel cancer diagnostics based on these genes. Its subsidiary, Rosetta Green (TASE:RSTG), was acquired by Monsanto for $35M. Under Isaac’s leadership, team members at Rosetta went on to lead artificial intelligence at IBM, Google and Microsoft. Now, at Quris, Isaac and his team are using a similar bio-AI approach to disrupt the drug development process.

The full potential of preventive medicine has yet to be unlocked. 
There are challenges in the way for physicians and health plans: it’s hard to identify the right members for valuable interventions, members are not engaging and the operation lacks precision, efficiency and effectiveness. 
As a result, members enter proactive clinical programs too late and avoidable health incidents are missed. 

Traditional solutions attempt to explain why patients are at risk. 
However, is it sufficient? For efficient preventive care one should identify which treatment is best for each patient, as well as it’s effectiveness.

In this talk, we will present a Proactive AI platform that identifies the most impactable members and provides actionable insights.
By advanced explanations and causal inference methodologies, and by utilizing multi-modal models, we provide for each patient at risk the link between specific clinical events and their impact and allow to assign each patient the right and most impactful intervention based on the AI-based actionable information.

Dr. Kira Radinsky is the CEO and CTO of Diagnostic Robotics, where the most advanced technologies in the field of artificial intelligence are harnessed to make healthcare better, cheaper, and more widely available.
In the past, she co-founded SalesPredict, acquired by eBay in 2016, and served as eBay director of data science and IL chief scientist.
One of the up-and-coming voices in the data science community, she is pioneering the field of medical data mining.
Dr. Radinsky gained international recognition for her work at Microsoft Research, where she developed predictive algorithms that recognized the early warning signs of globally impactful events, including political riots and disease epidemics. In 2013, she was named to the MIT Technology Review’s 35 Young Innovators Under 35, in 2015 as Forbes 30 under 30 rising stars in enterprise technology, and in 2016 selected as "woman of the year" by Globes. She is a frequent presenter at global tech events, including TEDx, Wired, Strata Data Science, Techcrunch and academic conferences, and she publishes in the Harvard Business Review.
Radinsky serves as a board member in: Israel Securities Authority, Maccabi Research Institute, and technology board of HSBC bank. 
Dr. Radinsky also serves as visiting professor at the Technion, Israel’s leading science and technology institute, where she focuses on the application of predictive data mining in medicine.
 

Large Language Models (LMs) are Improving at an incredible pace, but some inherent limitations of Transformer-based LMs remain problematic, e.g., generating factually ungrounded text, processing only bounded length text sequences, and more. In recent years, a multitude of successful solutions for these issues has been proposed. However, these involve specialized architectures and cannot be applied to off-the-shelf large LMs, rendering them irrelevant to models coming off the mainstream large-LM production line. In this talk I will present new methods for tackling the above drawbacks and others, which are applicable to any off-the-shelf large LM, without further training it. By addressing inherent LM weakspots in a model agnostic approach, we hope to alleviate them in a widespread manner, in existing and future leading LMs.

Yoav Levine serves as co-Chief Scientist in AI21 Labs, an Israeli start up in the field of NLP.  His academic research revolves around theoretical and practical aspects of using deep learning for natural language processing. His PhD is from the Hebrew University, and for it he has received the Israeli Academy of Sciences Adams fellowship and the Blavatnik PhD Prize, awarded annually to the top 5 Israeli PhD theses in the field of computer science. Prior to his doctoral studies, Yoav earned an M.Sc. in theoretical condensed matter physics from the Weizmann Institute, and a double B.Sc. in physics and electrical engineering (both summa cum laude) from Tel Aviv University, as a member of its Adi Lautman excellence program.

Incorporating humor in systems can enhance the user experience, attention and retention, as well as user’s rapport with the product. However, despite being studied for millennia, humor is computationally not well understood, often considered an AI-complete problem. Thus, much of the effort towards automating humor was done on very narrow and well-defined tasks. 
In this talk, I will present three novel computational humor works (and what they taught us):

In the first task we introduce a novel setting in humor mining: automatically detecting funny and unusual scientific papers [1]. We are inspired by the Ig Nobel prize, a satirical annual prize celebrating funny scientific achievements (example winner: “Chickens prefer beautiful humans”). 

In the second we explore playfulness in human-conversational agents (CAs) interactions [2]. While CAs are well-suited for task-oriented interactions (“Call Jason”), playful requests are often beyond their capabilities (“What’s your favorite color?”). This causes dissatisfaction and abandonment, destroying the rapport with the agent. Based on humor theories and real traffic data from Amazon Alexa, we present a taxonomy of playful requests and discuss ways to deploy it. 

In the third, we close the loop and explore ways to automate the process of generating appropriate responses to playful shopping requests [3].

We hope that insights gained from these studies will help tackle long-standing challenges in AI and social science.

[1] Shani, C., Borenstein, N., & Shahaf, D. (2021, August). How Did This Get Funded?! Automatically Identifying Quirky Scientific Achievements. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers) (pp. 14-28).

[2] Shani, C., Libov, A., Tolmach, S., Lewin-Eytan, L., Maarek, Y., & Shahaf, D. (2022, April). “Alexa, Do You Want to Build a Snowman?” Characterizing Playful Requests to Conversational Agents. In CHI Conference on Human Factors in Computing Systems Extended Abstracts (pp. 1-7).

[3] Shapira N., Kalinsky O., Libov A., Shani C. & Tolmach S. (2023). Evaluating Response Generation to Playful Shopping Requests. In European Conference on Information Retrieval. Springer, Cham.
 

Chen Shani is a Ph.D. student at the Hebrew university of Jerusalem, working with prof. Dafna Shahaf. Her prior education is in the fields of psychology and computational brain science.
She is interested in integrating insights from cognition to AI, and vice versa.
She worked on computational humor at Amazon Alexa for over three years. Nowadays she is working as an NLP consultant for startups.

In this talk Professor Eyal Zimlichman MD, MSC provides a review of how the future hospital must take shape through digital health tools utilized at ARC, the innovation arm of Sheba Medical Center (Sheba), ranked for four consecutive years as a top-10 medical center in the world by Newsweek. Professor Zimlichman will describe the Future of Health (FOH) community, which is a proactive forum of leaders working to influence the course of global health through creation of a shared vision. The ARC global initiative then works to turn the FOH vision into reality among the member organizations. To enable this transformation, ARC is based on four fundamental principles which include digital health, open & organic innovation, international collaborations, and a structured innovation methodology.
He continues with the variety of technologies developed and deployed at Sheba in artificial intelligence (AI), specifically in the field of digital radiology, digital pathology, AI decision support platform for clinical evidence, multi-omics, preventing medication errors, and an AI hospital wide control tower. Prof. Zimlichman concludes that all stakeholders must take a proactive approach in order to see the changes take shape earlier. These include policy makers, payers, care recipients, and caregivers who all need to be at the table to enable and deliver effective transformation.
 

Prof. Eyal Zimlichman is an internal medicine physician, a health care executive and a researcher focused on assessing and improving health care quality and value, patient engagement and patient safety. Prof. Zimlichman currently serves as Chief Transformation Officer and Chief Innovation Officer at Sheba Medical Center, Israel's largest hospital. In this position he oversees the transformation efforts carried out at the Medical Center as it aims to redesign healthcare to be focused on patients, while being more effective, efficient and equitable. For these reasons, Prof. Zimlichman founded Sheba’s global innovation program (ARC – Accelerate, Redesign and Collaborate). Prior to this Prof. Zimlichman held the position of Chief Medical Officer and before that as Chief Quality Officer, both at Sheba. Preceding his work at Sheba, Prof. Zimlichman held the position of Lead Researcher at Partners Health Care Clinical Affairs Department in Boston where he was involved in the efforts to bring about a strategic care redesign initiative. In that capacity, Prof. Zimlichman established their program for collecting and reporting patient reported outcomes across the continuum of care, a program that earned international appraisal. During that time, Prof. Zimlichman held an appointment at Brigham and Women’s Hospital and Harvard Medical School affiliated Center for Patient Safety Research and Practice, where he conducted research on implementing technology to improve quality and patient safety. Prof. Zimlichman is a founding member of the International Academy of Quality and Safety in Health Care and an International Expert at the International Society of Quality in Healthcare (ISQua). He also served as an advisor to the Office of the National Coordinator for Health Information Technology in the U.S. Department of Health and Human Services. In 2007 Prof. Zimlichman was selected as a member of Sheba's Talpiot Medical Leadership Program, a program aimed at advancing the brightest young physicians into leadership roles. In 2017 he was appointed head of the program. Prof. Zimlichman is a graduate of the Harvard School of Public Health Master of Science in Healthcare Management program and has earned his MD at the Technion Israel Institute of Technology.

Uri Rivner has been fighting financial crime for 20 years, working closely with the world's largest banks on developing strategies against online fraud. Prior to founding Refine Intelligence, Uri was co-founder and Chief Cyber Officer at BioCatch, the global leader in behavioral biometrics for fraud detection. Earlier Uri served as Head of New Technologies at security giant RSA. Innovations Uri spearheaded, such as Risk-Based Authentication and the world's first eFraud Network, are now saving the industry billions of dollars in fraud each year.

Ziv Freund joined Elbit Systems - EW & SIGINIT - Elisra in 2002 as an Algorithm Engineer in the field of computer vision. Ziv led Elisra's Computer Vision R&D group from 2012 to 2017. As of 2017, Ziv has been leading Elisra's AI R&D group, which focuses on applying artificial intelligence to different problems in signal processing, such as radar, communication and acoustic signals. Ziv holds a master's degree in Electrical Engineering and Business Administration from Tel Aviv University.