← Search

Avinash Baidya

2 accepted papers

2025

The Behavior Gap: Evaluating Zero-shot LLM Agents in Complex Task-Oriented Dialogs

ACL 2025finding

Large Language Model (LLM)-based agents have significantly impacted Task-Oriented Dialog Systems (TODS) but continue to face notable performance challenges, especially in zero-shot scenarios. While prior work has noted this performance gap, the behavioral factors driving the performance gap remain u…

2022

The Missing Invariance Principle found -- the Reciprocal Twin of Invariant Risk Minimization

NeurIPS 2022accept

Machine learning models often generalize poorly to out-of-distribution (OOD) data as a result of relying on features that are spuriously correlated with the label during training. Recently, the technique of Invariant Risk Minimization (IRM) was proposed to learn predictors that only use invariant fe…