BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//The Artificial Intelligence Materials Institute - ECPv6.15.19//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:The Artificial Intelligence Materials Institute
X-ORIGINAL-URL:https://aimi.cornell.edu
X-WR-CALDESC:Events for The Artificial Intelligence Materials Institute
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20250309T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20251102T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20260308T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20261101T060000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:20270314T070000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:20271107T060000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=America/New_York:20260507T150000
DTEND;TZID=America/New_York:20260507T160000
DTSTAMP:20260504T230140
CREATED:20260430T173839Z
LAST-MODIFIED:20260430T174028Z
UID:1667-1778166000-1778169600@aimi.cornell.edu
SUMMARY:AI-MI Seminar Series: Analyzing the Nonlocality of Sparse Autoencoder Features
DESCRIPTION:Join us for the next AI-MI Seminar Series talk as we explore how sparse autoencoders are helping unlock interpretable features within large language models. \nThis talk examines how tools from machine learning and theoretical physics can be combined to better understand the internal representations of LLMs. Drawing inspiration from holographic duality\, the speaker introduces a novel entropy-based measure to quantify how nonlocal learned features are in relation to input tokens—offering new insight into how information is structured and processed in these systems. \nWatch live at youtube.com/@AIMaterialsInstitute \nTopic: Sparse autoencoders (SAEs) have emerged as a useful tool for extracting interpretable features from the internal representations of large language models (LLMs). Motivated by an analogy with holographic duality in theoretical physics\, where strongly correlated boundary degrees of freedom map to weakly coupled bulk fields of varying nonlocality\, we introduce an entropy measure that quantifies how nonlocal each SAE feature is in the input token space\, in the sense of how many input tokens have a strong impact on the activation of this feature. We analyze this measure in different language models\, providing a new tool for understanding the information dynamics in LLMs. \nSpeaker: Xiaoliang Qi is a Professor of Physics at Stanford University\, where his research spans the interplay of quantum entanglement\, quantum gravity\, and quantum chaos\, alongside continued work on topological phases in condensed matter systems. He is a recipient of the New Horizons in Physics Prize and the Packard Fellowship\, among other honors. Qi has been a leading voice in applying ideas from quantum information and tensor networks to many-body and materials problems\, and is engaged in the emerging conversation about agentic and AI-driven approaches to scientific research.
URL:https://aimi.cornell.edu/event/ai-mi-seminar-series-analyzing-the-nonlocality-of-sparse-autoencoder-features/
CATEGORIES:Seminar
ATTACH;FMTTYPE=image/jpeg:https://aimi.cornell.edu/wp-content/uploads/YouTube-event-listing-THUMBNAIL-1-1.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;VALUE=DATE:20260519
DTEND;VALUE=DATE:20260521
DTSTAMP:20260504T230140
CREATED:20260105T032736Z
LAST-MODIFIED:20260121T190141Z
UID:1007-1779148800-1779321599@aimi.cornell.edu
SUMMARY:AI-MI Annual Meeting
DESCRIPTION:Members of the NSF AI Materials Institute will convene for the AI-MI Annual Meeting\, a two-day\, in-person meeting of the NSF AI Materials Institute community. The meeting will feature research updates across the institute\, student and trainee spotlights\, focused discussions on shared challenges and opportunities\, and dedicated time for networking and collaboration. We welcome faculty\, staff and students to connect\, align on priorities\, and accelerate progress toward AI-enabled materials discovery.
URL:https://aimi.cornell.edu/event/ai-mi-annual-meeting/
END:VEVENT
END:VCALENDAR