Leonid Sigal

SPIKE-RL: Video-LLMs meet Bayesian Surprise

To Sink or Not to Sink: Visual Information Pathways in Large Vision-Language Models

Free Lunch Alignment of Text-to-Image Diffusion Models without Preference Image Pairs

Motion-Aware Surface Smoothing for Monocular Avatar Representations

ADiff4TPP: Asynchronous Diffusion Models for Temporal Point Processes

Learning What Matters: Prioritized Concept Learning via Relative Error-driven Sample Selection

SCALE-VLP: Soft-Weighted Contrastive Volumetric Vision–Language Pre-training with Spatial-Knowledge Semantics

RewardRank: Optimizing True Learning-to-Rank Utility

Can Multi-Modal LLMs Provide Live Step-by-Step Task Guidance?

AoPS Dataset: Leveraging Online Olympiad-Level Math Problems for LLMs Training and Contamination-Resistant Evaluation

Locality Sensitive Avatars From Video

Revealing Weaknesses in Text Watermarking Through Self-Information Rewrite Attacks

MM-R$^3$: On (In-)Consistency of Multi-modal Large Language Models (MLLMs)

On Pre-training of Multimodal Language Models Customized for Chart Understanding

Leveraging Online Olympiad-Level Math Problems for LLMs Training and Contamination-Resistant Evaluation

SIRA: Exposing Vulnerabilities in Text Watermarking with Self-Information Rewrite Attacks