Companies & labs
MIT
Massachusetts Institute of Technology, US
4recordsFirst record 2023Latest Sep 29, 2023
Records 4
Newest first. Open one to see it with its source.
MIT (Xiao et al.)
StreamingLLM
The finding that initial tokens act as attention sinks; 2 years later gpt-oss adopted learnable sinks
MIT Han Lab (Ji Lin et al.)
AWQ
Quantization that protects only the top weights as judged by activations; TinyChat runs more than 3× faster than HF FP16
Yilun Du et al. (MIT CSAIL)
Multiagent Debate published
The reference paper for the multi-agent debate line of work
Prompt EngineeringPaper
Noah Shinn et al. (Northeastern University, MIT)
Shinn et al. present Reflexion
The lineage of verbal reinforcement learning without weight updates. The v1 title was an autonomous agent with dynamic memory and self-reflection
Prompt EngineeringPaper