vlm

Personal update on hardware water damage recovery, showcasing MLX-VLM serving Qwen3-4B-Instruct locally on an RTX6000 Pro at ~300 tok/s for autocomplete and git commit generation via Zed IDE.

MemEye is a visual-centric evaluation framework that assesses multimodal agent memory by measuring visual evidence granularity and retrieval complexity across 8 life-scenario tasks, revealing that current architectures struggle to preserve fine-grained visual details and reason about state changes over time.

This paper introduces FragileFlow, a plug-in regularizer that improves the robustness of LLMs and VLMs by controlling 'correct-but-fragile' predictions through spectral analysis and PAC-Bayes bounds.

This survey paper introduces World Action Models (WAMs), a unified framework for embodied AI that integrates predictive state modeling with action generation. It provides a taxonomy of existing methods, analyzes the data ecosystem, and outlines evaluation protocols for this emerging paradigm.

This paper introduces the Auto-Rubric as Reward (ARR) framework, which externalizes implicit preference knowledge into explicit rubrics for multimodal alignment. It proposes Rubric Policy Optimization (RPO) to stabilize policy gradients, achieving better performance in text-to-image and image editing tasks.

ParseBench introduces the first benchmark evaluating vision-language models on chart comprehension within full enterprise documents, addressing gaps in prior chart-only benchmarks.

NomadicAI is building an agentic computer-vision product to fix VLMs' weak grounding in actual video content.

Jerry Liu discusses challenges with using Vision Language Models for PDF parsing, particularly around ensuring text correctness and maintaining proper reading order while avoiding hallucinations.

PersonaVLM introduces a personalized multimodal LLM framework that enables long-term user adaptation through memory retention, multi-turn reasoning, and response alignment, outperforming GPT-4o by 5.2% on the new Persona-MME benchmark.