Leonid Karlinsky

Visualizing Thought: Conceptual Diagrams Enable Robust Planning in LMMs

Activation Reward Models for Few-Shot Model Alignment

Overflow Prevention Enhances Long-Context Recurrent LLMs

Self-MoE: Towards Compositional Large Language Models with Self-Specialized Experts

GLOV: Guided Large Language Models as Implicit Optimizers for Vision Language Models

BAT-CLIP: Bimodal Test-Time Adaptation for CLIP

LiveXiv - A Multi-Modal live benchmark based on Arxiv papers content

Comparison Visual Instruction Tuning

Navigating the Labyrinth: Evaluating and Enhancing LLMs’ Ability to Reason About Search Problems