Tagged articles

Multimodal Large Language Models

4 articles · Page 1 of 1

Oct 17, 2026 · Artificial Intelligence

RAVEL: Two Simple Tweaks Double Vision Encoder Evaluation Correlation for MLLMs

Researchers propose RAVEL, a training-free method to evaluate vision encoders for multimodal LLMs by measuring cross-modal nearest-neighbor consistency with stabilized PCA whitening and patch-level matching, achieving 0.85 Spearman correlation across 210 models, outperforming prior methods by 36-101%.

Multimodal Large Language ModelsPCA WhiteningPatch-Level Matching

0 likes · 13 min read

RAVEL: Two Simple Tweaks Double Vision Encoder Evaluation Correlation for MLLMs

Kuaishou Tech

Jul 6, 2026 · Artificial Intelligence

ICML 2026 Spotlight: MetaphorVU – The First Benchmark for Metaphorical Video Understanding

The MetaphorVU project introduces the first systematic benchmark for metaphor video understanding, builds a taxonomy of eight metaphor types from billions of real short videos, evaluates 11 multimodal LLMs revealing a 20‑point gap to human performance, and proposes MetaphorBoost—a knowledge‑graph‑enhanced inference framework that consistently improves metaphor comprehension across models.

ICML 2026Knowledge GraphMetaphorBoost

0 likes · 14 min read

ICML 2026 Spotlight: MetaphorVU – The First Benchmark for Metaphorical Video Understanding

Xiaomi Tech

Mar 3, 2026 · Artificial Intelligence

Xiaomi Scores 14 Papers at CVPR 2026, Showcasing Breakthroughs in Large Models and Autonomous Driving

CVPR 2026 accepted 14 Xiaomi papers spanning long‑video understanding, multimodal reasoning, GUI agents, and autonomous driving, each accompanied by arXiv and GitHub links, and introducing novel frameworks such as REVISOR, EMO‑R3, TimeViper, MSJoE, SafeGRPO, GUI‑CEval, ProactiveMobile, ParkGaussian, UFO, TraqPoint, SimScale, MeanFuser and DVGT.

Autonomous DrivingCVPR 2026Long Video Understanding

0 likes · 19 min read

Xiaomi Scores 14 Papers at CVPR 2026, Showcasing Breakthroughs in Large Models and Autonomous Driving

JD Tech Talk

Mar 13, 2025 · Artificial Intelligence

CTR-Driven Advertising Image Generation with Multimodal Large Language Models

This paper proposes CAIG, a novel method for generating high-CTR advertising images using multimodal large language models, combining reinforcement learning and preference optimization to align generated content with product features.

CTR predictionMultimodal Large Language Modelsadvertising image generation

0 likes · 10 min read

CTR-Driven Advertising Image Generation with Multimodal Large Language Models