Dynamic Steering of Layer-wise Uncertainty to Mitigate Hallucination in LVLMs
News
- One paper on spatial reasoning accepted at EMNLP 2026 Findings
- Two papers on logical and spatial reasoning accepted at ACL 2026 Main
- One paper on video anomaly detection accepted at ICPR 2026
- Three papers accepted at ICLR 2026, EACL 2026 Findings, and PAKDD 2026
- One paper on LVLM hallucination accepted at ICCV 2025
Publications
* equal contribution
Under review
BDI Architecture for Efficient Orchestration of LLM-based Multi-Agent Systems
Failure Attribution with Stepwise Uncertainty in Multi-Agent Systems
Published
Spatial Reasoning in Large Vision-Language Models via Egocentric-to-Allocentric Transformation
Diagnosing Spatial Consistency across Perspectives and Viewpoints in Large Vision-Language Models
Hyperbolic Spatio-temporal Representation Learning for Unsupervised Video Anomaly Detection
Adaptive Beam Search with Shannon Entropy for Data-centric Reasoning in LLMs
Visual Question Answering: A Survey of Methods, Datasets, Evaluation, and Challenges
No papers under this topic yet.
Earlier work
-
FTT: Fourier Transform based Transformer for Brain CT Report Generation
-
Facial Expression Recognition using Visual Transformer with Histogram of Oriented Gradients
-
Facial Expression Recognition Robust to Occlusion using Spatial Transformer Network with Triplet Loss Function
Beyond research
Outside the lab, I love dogs, DJing, and watching movies.