BEGIN:VCALENDAR
VERSION:2.0
CALSCALE:GREGORIAN
PRODID:adamgibbons/ics
METHOD:PUBLISH
X-PUBLISHED-TTL:PT1H
BEGIN:VEVENT
UID:EB5hi_o7pFz-qVNaFcVTc
SUMMARY:Scaling LLM-RL for the age of agents
DTSTAMP:20260430T144232Z
DTSTART:20260522T114500Z
DESCRIPTION:Description:\nReinforcement learning (RL) has emerged as the pr
	imary paradigm for scaling base models into autonomous agents across softw
	are engineering\, research\, and beyond. Yet scaling RL poses fundamentall
	y different challenges than pre-training. Its asynchronous and online natu
	re introduces new problems\, from curriculum design and synthetic task gen
	eration to off-policy training dynamics and large-scale sandboxed rollout 
	execution.In this talk\, Konstantin Dunas traces how RL scaling has evolve
	d from early approaches and why this shift creates new infrastructural bot
	tlenecks. He will examine challenges such as training stability\, training
	–inference policy mismatch\, and off-policy learning\, and how prime-rl\, 
	Prime Intellect’s open-source training framework\, enables others to apply
	 these approaches to train their own agents. The talk will also outline wh
	at comes next for large-scale LLM-RL.\n--------------------------------\n\
	nSpeaker:\n- Konstantin Dunas\n\n--------------------------------\n\nTalk 
	details:\n- Link to the Big Techday website: https://bigtechday.com/en/tal
	ks#7Bzb87sZbHksbnpcYOf6fK\n
LOCATION:Stellwerk
DURATION:PT50M
END:VEVENT
END:VCALENDAR