BEGIN:VCALENDAR
PRODID:-//AddEvent Inc//AddEvent.com v1.7//EN
VERSION:2.0
BEGIN:VTIMEZONE
TZID:America/Los_Angeles
BEGIN:STANDARD
DTSTART:20261101T010000
RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
TZNAME:PST
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20260308T030000
RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
TZNAME:PDT
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
DESCRIPTION:This talk explores a surprising finding in reinforcement learning for mathematical reasoning: certain language models can achieve substantial performance gains even when trained with completely uninformative or incorrect reward signals. Through extensive experiments on mathematical benchmarks\, we demonstrate that Qwen2.5-Math models improve significantly when trained with random rewards\, format-only rewards\, or even rewards that explicitly favor incorrect answers.\n\nStella a second year Ph.D. student in the Allen School of Computer Science and Engineering at the University of Washington\, advised by Yulia Tsvetkov.\n\nStella received her B.S. and M.S.E. at Johns Hopkins with majors in Computer Science\, Cognitive Science (linguistics focus)\, and Applied Mathematics (statistics focus). I worked as a research assistant at the Center for Language and Speech Processing advised by Philipp Koehn and Kenton Murray.\n\n------\n\nCreate your own Add to Calendar links with addevent.com/r/a \n
X-ALT-DESC;FMTTYPE=text/html:This talk explores a surprising finding in reinforcement learning for mathematical reasoning: certain language models can achieve substantial performance gains even when trained with completely uninformative or incorrect reward signals. Through extensive experiments on mathematical benchmarks, we demonstrate that Qwen2.5-Math models improve significantly when trained with random rewards, format-only rewards, or even rewards that explicitly favor incorrect answers.<br><br>Stella a second year Ph.D. student in the Allen School of Computer Science and Engineering at the University of Washington, advised by Yulia Tsvetkov.<br><br>Stella received her B.S. and M.S.E. at Johns Hopkins with majors in Computer Science, Cognitive Science (linguistics focus), and Applied Mathematics (statistics focus). I worked as a research assistant at the Center for Language and Speech Processing advised by Philipp Koehn and Kenton Murray.<br /><br />------<br /><br />Create your own Add to Calendar links with addevent.com/r/a <br>
UID:cc6caa5f291e4ab394a2cc56827f012aaddeventcom
SUMMARY:Stella Li - Spurious Rewards: Rethinking Training Signals in RLVR (Geo Asia)
DTSTART;TZID=America/Los_Angeles:20250611T090000
DTEND;TZID=America/Los_Angeles:20250611T100000
DTSTAMP:20261008T231521Z
TRANSP:OPAQUE
STATUS:CONFIRMED
SEQUENCE:0
LOCATION:https://meet.google.com/yhv-tiir-ava?hs=122&authuser=0
X-MICROSOFT-CDO-BUSYSTATUS:BUSY
BEGIN:VALARM
TRIGGER:-PT30M
ACTION:DISPLAY
DESCRIPTION:Reminder
END:VALARM
END:VEVENT
END:VCALENDAR