Harsha Vardhan Khurdula's picture

Harsha Vardhan Khurdula

HV-Khurdula

·

AI & ML interests

Transformers, Vision-Transformers, Multi-Agent Frameworks, Datatset Curation & Benchmarking.

Recent Activity

authored a paper 13 days ago

Interfaze: The Future of AI is built on Task-Specific Small Models

authored a paper 14 days ago

The Structured Output Benchmark: A Multi-Source Benchmark for Evaluating Structured Output Quality in Large Language Models

upvoted a paper 14 days ago

Tuna-2: Pixel Embeddings Beat Vision Encoders for Multimodal Understanding and Generation

View all activity

Organizations

upvoted a paper 14 days ago

Tuna-2: Pixel Embeddings Beat Vision Encoders for Multimodal Understanding and Generation

Paper • 2604.24763 • Published 16 days ago • 70

upvoted a paper 22 days ago

Interfaze: The Future of AI is built on Task-Specific Small Models

Paper • 2602.04101 • Published Feb 4 • 4

upvoted 2 papers 8 months ago

Scrub It Out! Erasing Sensitive Memorization in Code Language Models via Machine Unlearning

Paper • 2509.13755 • Published Sep 17, 2025 • 19

Revolutionizing Reinforcement Learning Framework for Diffusion Large Language Models

Paper • 2509.06949 • Published Sep 8, 2025 • 57

upvoted an article 9 months ago

Article

From Zero to GPU: A Guide to Building and Scaling Production-Ready CUDA Kernels

drbh, danieldk

•

Aug 18, 2025

• 97

upvoted 2 papers 10 months ago

Scaling RL to Long Videos

Paper • 2507.07966 • Published Jul 10, 2025 • 161

OST-Bench: Evaluating the Capabilities of MLLMs in Online Spatio-temporal Scene Understanding

Paper • 2507.07984 • Published Jul 10, 2025 • 43