I am actively interested in talking about industry research scientist and engineering positions in model evaluation and benchmarking, metric design, post-training alignment, model explanation, and interpretation, and, work at the intersection of LLMs and productivity or health (see publications Show information for the linked content and research notes). If you have any in mind, please reach out to me.
I am also interested in collaborations in the aforementioned areas. If you have an interesting project, and would like to collaborate or talk, or just want to talk about research in general, please pick a time slot hereShow information for the linked content .
Hi, I am Mimansa
I’m a member of technical staff at Norm AI. I obtained my doctorate in CS at the University of Michigan under Prof. Emily Provost as part of the CHAI group. My research focuses on developing cost-effective data collection and generation procedures, LLM orchestration schemas, and, designing evaluation methods and metrics that are interpretable and human aligned.
I've previously interned at Allen AI NLP, Facebook AI Research (FAIR) NLP, and Facebook Research Conversational AI.
Besides research, I am interested in science communication, sketchnoting, personal knowledge management and cooking.
I ❤️ cats, and they keep me sane. I have two: OreoShow information for the linked content and BertShow information for the linked content (yes, that Bert, you read it right!).
Experience
-
Oct 2023 - Present
Member of Technical Staff - Research Scientist, Norm AI
Retrieval Augmentation and persona based simulation in LLMs and their subjective evaluation -
Fall 2021
Research Intern, Allen NLP, Allen Institute of Artificial Intelligence
Mentor(s) : Ana Marasović
Benchmarking interpretable and compositional evaluation for very large models in natural language processing -
Summer 2021
Research Intern, NLP Team, Facebook AI Research (Meta AI)
Mentor(s) : Adina Williams; Scott Yih, Pedro Rogieruez
Identifying, understanding, and correcting ambiguity based faliure points in Natural Language Inference (NLI) -
Summer 2020
Research Intern, Conversation AI, Facebook (Meta) AI
Mentor(s) : Ahmad Beirami, Shane Moon, Satwik Kottur, Chinnadhurai Sankar
Defining interpretable and generalizable user satisfaction metric informed by human knowledge -
Summer 2016
Research Intern, SENTIC Lab, Nanyang Technological University, Singapore
Mentor(s) : Erik Cambria
Multimodal detection of human behavior: empathy and deception -
Winter 2016
Research Intern, NLP and Sentiment Analysis (NLPSA) Lab, Academia Sinica, Taiwan
Mentor(s) : Lun-Wei Ku
Human perception of conversation sentiment visualization in chat interfaces -
Summer 2015
Research Intern, Indraprastha Institute of Information Technology (IIIT), Delhi
Mentor(s) : Pravesh Biyani
Finding the optimal routes under user specified user-constraints for last mile cab-pooling
Education
-
(2017 - 2023)
PhD in Computer Science and Engineering
University of Michigan
Advisor(s): Prof. Emily Provost -
(2017 - 2019)
MS in Computer Science and Engineering
University of Michigan
Advisor(s): Prof. Emily Provost -
(2013 - 2017)
BTech in Computer Engineering
IET DA Univ, India
Advisor(s): Prof. G.L. Prajapati