6 papers
Engineering Systems for Data Analysis Using Interactive Structured Inductive Programming
Shraddha Surana, Ashwin Srinivasan, Michael Bain
Engineering information systems for scientific data analysis presents significant challenges: complex workflows requiring exploration of large solution spaces, close collaboration…
Multi-Turn Human-LLM Interaction Through the Lens of a Two-Way Intelligibility Protocol
Harshvardhan Mestha, Karan Bania, Shreyas V Sathyanarayana +2
Our interest is in the design of software systems involving a human-expert interacting -- using natural language -- with a large language model (LLM) on data analysis tasks. For co…
Agent-Based Detection and Resolution of Incompleteness and Ambiguity in Interactions with Large Language Models
Riya Naik, Ashwin Srinivasan, Swati Agarwal +1
Many of us now treat LLMs as modern-day oracles asking it almost any kind of question. However, consulting an LLM does not have to be a single turn activity. But long multi-turn in…
Structured Program Synthesis using LLMs: Results and Insights from the IPARC Challenge
Shraddha Surana, Ashwin Srinivasan, Michael Bain
The IPARC Challenge, inspired by ARC, provides controlled program synthesis tasks over synthetic images to evaluate automatic program construction, focusing on sequence, selection,…
An Empirical Study of the Role of Incompleteness and Ambiguity in Interactions with Large Language Models
Riya Naik, Ashwin Srinivasan, Estrid He +1
Natural language as a medium for human-computer interaction has long been anticipated, has been undergoing a sea-change with the advent of Large Language Models (LLMs) with startli…
Concept Distillation from Strong to Weak Models via Hypotheses-to-Theories Prompting
Emmanuel Aboah Boateng, Cassiano O. Becker, Nabiha Asghar +5
Hand-crafting high quality prompts to optimize the performance of language models is a complicated and labor-intensive process. Furthermore, when migrating to newer, smaller, or we…