Select a calendar:
Filter July Events by Event Type:
SUNMONTUEWEDTHUFRISAT
Events for July 12, 2024
-
CS Seminar: Michael Pradel (University of Stuttgart) - Neuro-Symbolic Developer Tools for Analyzing, Executing, and Repairing Code
Fri, Jul 12, 2024 @ 11:00 AM - 12:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Michael Pradel, University of Stuttgart
Talk Title: Neuro-Symbolic Developer Tools for Analyzing, Executing, and Repairing Code
Abstract: Developer productivity and software quality critically depend on effective software development tools. Traditional, symbolic program analysis tools are often limited in their ability to understand developer intention and rely on various hand-crafted heuristics. Neural software analysis addresses these limitations, but remains unaware of the formal semantics of a program and hence easily misses facts and rules that are actually well known. This talk argues that carefully combining neural and symbolic reasoning provides an effective means to address various challenging software development problems. To illustrate this point, I will describe our 8-year long journey of creating neuro-symbolic developer tools, ranging from learning-based bug detectors and type predictors, to our most recent work on learning-guided execution and program repair based on an autonomous LLM-based agent. I will discuss lessons learned on this journey and conclude with an outline of open challenges waiting to be addressed in order to close the gap between symbolic and neural software developer tools.The talk is based (mostly) on these papers:https://arxiv.org/abs/2302.02343 https://arxiv.org/abs/2403.17134
Biography: Michael Pradel is a full professor at the University of Stuttgart, which he joined after a PhD at ETH Zurich, a post-doc at UC Berkeley, an assistant professorship at TU Darmstadt, and a sabbatical at Facebook. His research interests span software engineering, programming languages, security, and machine learning, with a focus on tools and techniques for building reliable, efficient, and secure software. In particular, he is interested in neural software analysis, analyzing web applications, dynamic analysis, and test generation. Michael has been recognized through the Ernst-Denert Software Engineering Award, an Emmy Noether grant by the German Research Foundation (DFG), an ERC Starting Grant, best/distinguished paper awards at FSE (3x), ISSTA, ASE, and ASPLOS, and by being named an ACM Distinguished Member.
Host: Chao Wang
Location: Ronald Tutor Hall of Engineering (RTH) - 115
Audiences: Everyone Is Invited
Contact: CS Faculty Affairs
-
AI Seminar- Nexa AI – Functional Tokens for On-device Multimodal Models
Fri, Jul 12, 2024 @ 11:00 AM - 12:00 PM
Information Sciences Institute
Conferences, Lectures, & Seminars
Speaker: Alex Chen, CEO + Founder of Nexa AI and Zack Li, CTO + Co-Founder of Nexa AI, Nexa AI
Talk Title: Nexa AI -“ Functional Tokens for On-device Multimodal Models
Series: AI Seminar
Abstract: Zoom meeting ID: 944 0958 4905Passcode: 822247
Tokenizing corpora into semantic tokens has proven effective for large language models. However, this approach encounters challenges when applied to function calls, leading to inaccuracies and hallucinations. To address this issue, we have pioneered a new training methodology using functional tokens, transforming complex function calling tasks into language completion tasks. We also released Octopus-series models using functional tokens and achieved GPT4 level function calling accuracy with 2B parameter size. Our Octopus-V2 model achieved 35 times faster inference speed up and 70 times more energy efficiency compared to the RAG plus Llama3 solution, and is four times faster than OpenAI’s GPT-4O. The functional token is then applied to Octopus-V3, a sub-billion multimodal model, adept at both text and images, and fluent in English and Mandarin. Furthermore, Octopus-V4 extends these capabilities into a graph network structure, with Octopus-V2 as the master node and integration with other open-source models as worker nodes, Octopus-V4 achieved 74.8 MMLU and outperforms GPT3.5, and applied for cloud and edge collaboration. Nexa’s Octopus-V2 models ranked 2nd place among half a million models on HuggingFace between Apr 2 and Apr 15, surpassing XAI grok and Databrick DBRX model during that period, and was mentioned by Google Gemma team during the 2024 Google IO. Nexa’s Octopus models have also attracted industrial collaboration interest from AWS, Google, Volkswagen US, Qualcomm, ByteDance, Stellantis, Zoom, and more.
Biography: Alex Chen is the CEO and founder of Nexa AI, with PhD in Mechanics and Computation from Stanford University. His research interests lie in AI agent development empowered by large language models. He is a serial entrepreneur and served as President of the Chinese Entrepreneur Organization before. He is also a gold medalist in the Mathematics Olympiad. Zack Li is the CTO and co-founder of Nexa AI. Before this, he accumulated four years of industrial experience in on-device AI at Google and Amazon Lab126, focusing on model deployment, performance optimization, and edge-cloud collaboration. He received an MS in Operation Research from Stanford University. Alex and Zack are founders of Nexa AI and have authored Octopus series models. Nexa AI builds lightweight but powerful multimodal models for AI agents and provides on-device SDK infra to make models run fast and energy-efficiently. For more information, visit https://www.nexa4ai.com/
If speaker approves to be recorded for this AI Seminar talk, it will be posted on our USC/ISI YouTube page within 1-2 business days: https://www.youtube.com/user/USCISI.
Subscribe here to learn more about upcoming seminars: https://www.isi.edu/events/
Host: Abel Salinas and Justina Gilleland
More Info: https://www.isi.edu/events/5009/nexa-ai-functional-tokens-for-on-device-multimodal-models/
Webcast: https://usc.zoom.us/j/94409584905?pwd=Sm5LVkd0bndUdEluM3piK0NWTUQrUT09Location: Information Science Institute (ISI) - Virtual Only
WebCast Link: https://usc.zoom.us/j/94409584905?pwd=Sm5LVkd0bndUdEluM3piK0NWTUQrUT09
Audiences: Everyone Is Invited
Contact: Pete Zamar
Event Link: https://www.isi.edu/events/5009/nexa-ai-functional-tokens-for-on-device-multimodal-models/
