BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//OneJoy//events//EN
BEGIN:VEVENT
UID:onejoy-event-18501@onejoy
DTSTAMP:20261005T095526Z
DTSTART:20260825T003000Z
DTEND:20260825T023000Z
SUMMARY:Gen AI Paper Reading: ub 1-bit Quantization with Nanobit and Littlebitv2
LOCATION:
DESCRIPTION:Join us for a paper discussion on "NANOQUANT: Efficient Sub-1-Bit Quantization of Large Language Models" and "LittleBit-2: Maximizing the Spectral Energy Gain in Sub-1-Bit LLMs\nvia Latent Geometry Alignment" presented by Logan. These papers cover the latest developments in llm quantization where sub 1-bit llm quantization are achieved through combination of sparse matrix approaches and more traditional quantization methods.\nlittlebitv2 (https://arxiv.org/pdf/2603.00042)\nnanoquant (https://arxiv.\nhttps://onejoy.io/event/18501
URL:https://onejoy.io/event/18501
END:VEVENT
END:VCALENDAR