Chitta Baral

GuidedSampling: Steering LLMs Towards Diverse Candidate Solutions at Inference-Time

Build-Bench: Benchmarking LLM Agents on Compiling Real-World Open Source Software

Chimera: Compositional Image Generation using Part-based Concepting

Investigating VLM Hallucination from a Cognitive Psychology Perspective: A First Step Toward Interpretation with Intriguing Observations

EraseFlow: Learning Concept Erasure Policies via GFlowNet-Driven Alignment

Stable Cinemetrics : Structured Taxonomy and Evaluation for Professional Video Generation

ActionReasoningBench: Reasoning about Actions with and without Ramification Constraints

VOILA: Evaluation of MLLMs For Perceptual Understanding and Analogical Reasoning

PolyMATH: A Challenging Multi-Modal Mathematical Reasoning Benchmark

Dual Caption Preference Optimization for Diffusion Models