iiinova — where your digital self keeps evolving

Priyanka Kargupta @ EMNLP'24

@pkargupta.bsky.social

109 followers 291 following 6 posts

structured nlp to help models & humans critically think nlp phd candidate @UIUC_NLP | nsf grfp fellow prev: cs'22 @berkeley_ai | @IntelAI | @IBMResearch

Posts Replies Media

Priyanka Kargupta @ EMNLP'24 @pkargupta.bsky.social · 26/11/2024

hello! would appreciate being added, thank you!

010

Priyanka Kargupta @ EMNLP'24 @pkargupta.bsky.social · 19/11/2024

Hello! Would love to be added to this, thank you so much!

110

Priyanka Kargupta @ EMNLP'24 @pkargupta.bsky.social · 19/11/2024

TreeInstruct is preferred 78.43% of the time. It solves 14.09% more bugs across all settings, and our questions are 14.18% better at addressing bugs, maintaining relevance, and ensuring logical conversation flow. TreeInstruct also adapts to human students of varying backgrounds.

010

Priyanka Kargupta @ EMNLP'24 @pkargupta.bsky.social · 19/11/2024

TreeInstruct estimates the knowledge a student needs to debug their code and devises a conversation plan. It then dynamically constructs a question tree based on its interactions with the student, navigating the knowledge state space till the student comprehends & fixes all misunderstandings &bugs.

110

Priyanka Kargupta @ EMNLP'24 @pkargupta.bsky.social · 19/11/2024

github.com/agarwalishik... We apply TreeInstruct to code debugging. Prior works directly give away bugs/fixes, assume single-turn conversations, and only work for one bug. We create a realistic, multi-bug dataset, where the bugs can be either mutually independent or dependent.

110

Priyanka Kargupta @ EMNLP'24 @pkargupta.bsky.social · 19/11/2024

Can LLMs make us critical thinkers? TreeInstruct reorients assistant-like LLMs to be instructors that guide students towards understanding their mistakes, without providing direct/indirect answers. Check out aclanthology.org/2024.finding... (w/ @wonderingishika.bsky.social) to learn more!

121