BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//wp-events-plugin.com//7.4.0.1//EN
TZID:America/New_York
X-WR-TIMEZONE:America/New_York
BEGIN:VEVENT
UID:890@fds.yale.edu
DTSTART;TZID=America/New_York:20251020T153000
DTEND;TZID=America/New_York:20251020T170000
DTSTAMP:20251016T201135Z
URL:https://fds.yale.edu/events/sds-seminar-adam-block-columbia/
SUMMARY:S&amp\;DS Seminar: Adam Block (Columbia)\, "Scaling Inference-Time 
 Compute: From Self-Improvement to Pessimism"
DESCRIPTION:\nAbstract: Language models increasingly rely on scaling infere
 nce-time computation to achieve state-of-the-art performance on a growing 
 number of reasoning tasks.  A popular paradigm for such computational sca
 ling is Best-of-N (BoN) sampling\, where a model generates multiple candid
 ate responses to a given question and selects the one among them as the mo
 st likely to be correct.  In this talk I will present a unified understan
 ding of this approach in several settings\, both with and without external
  verification.  We will discuss the extent to which such inference-time c
 omputation is necessary as well as present a new algorithm that optimally 
 leverages inference-time compute to return better answers in the presence 
 of uncertainty\, thereby avoiding common pitfalls of BoN sampling such as 
 reward-hacking and over-optimization.  Throughout\, we will see that mode
 l coverage of ‘good’ answers emerges as the critical feature allowing 
 for inference-time computation to scale effectively.  These results provi
 de a principled foundation for designing inference-time algorithms that sc
 ale reliably with compute and highlight coverage as the central bottleneck
  in aligning language models.\n
CATEGORIES:FDS Events,Statistics & Data Science Seminar
LOCATION:Yale Institute for Foundations of Data Science\, Kline Tower 13th 
 Floor\, Room 1327\, New Haven\, CT\, 06511\, United States
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=Kline Tower 13th Floor\, Ro
 om 1327\, New Haven\, CT\, 06511\, United States;X-APPLE-RADIUS=100;X-TITL
 E=Yale Institute for Foundations of Data Science:geo:0,0
END:VEVENT
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:DAYLIGHT
DTSTART:20250309T030000
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
END:DAYLIGHT
END:VTIMEZONE
END:VCALENDAR