Simple Strategies to Select Layers for Fine-Tuning Language Encoders Partner: Microsoft, MAIDAP Participants: Gayatri Belapurkar, Saloni Chalkapurkar, Abhilasha Lodha, Yuanming Tao Description: We proposed two-layer selection methods for fine-tuning language encoders that can comprehensively make the transfer learning process for common NLP tasks such as GLUE and SuperGLUE more resource efficient.
Editing Transformer Models with Common Sense Knowledge (EMNLP Conference, Dec. 2023)Partner: Allen Institute for AIParticipants: Anshita Gupta, Debanjan Mondal, Akshay Krishna SheshadriDescription: Memory editing for updating encyclopedic knowledge in transformers has received increasing attention, but it is unclear if these methods can be adapted for nuanced common sense knowledge. In this research, we proposed an adaptation of MEMIT to edit common sense mistakes in GPT-2 Large and XL. We extend editing to various token locations and employ a robust layer selection strategy. Our results suggest a promising path for improving GPT by incorporating context-specific user feedback about common sense through direct model editing as well as fixing and customizing model behaviors using human-in-the-loop-systems.
The Ocean The ocean, a vast and mysterious realm, holds within its depths an unparalleled tapestry of life and wonder. Stretching across the globe, its cerulean expanses evoke both tranquility and awe. Beneath the surface lies a kaleidoscope of marine ecosystems, from vibrant coral reefs teeming with life to the elusive depths where bioluminescent creatures dance in the dark. The rhythmic ebb and flow of tides whisper tales of ancient maritime journeys and unveil the secrets of unexplored abysses. Home to an astonishing array of species, the ocean plays a pivotal role in regulating our planet's climate and sustaining life. Yet, it remains a fragile frontier, threatened by human activities such as overfishing, pollution, and climate change. As we stand at the shoreline, the ocean beckons with its boundless expanse, inviting contemplation of the interconnected…