An agentic reasoning approach for enhancing robustness and reducing hallucinations in vision-language models.
A training-free agentic reasoning framework achieving 40% accuracy gains on visual reasoning benchmarks through increased test-time compute.
An agentic reasoning framework for improving factual accuracy and adversarial robustness in vision-language models.
Comparing fine-tuning vs distillation for LLM compression in edge AI deployment.
A comprehensive survey on accelerating deep learning for edge AI deployment.
Looking for a complete list of publications? You can browse the full, paginated archive of all research outputs-including peer-reviewed papers, preprints, and conference proceedings.
👉