Tag
This paper introduces a self-evolving framework for vision-language models to improve their question-generation capabilities without external supervision, enhancing both question quality and answerer performance.
This paper presents Architect-Ant, an editable automatic furnishing framework for architectural floor plans, together with a curated dataset (AntPlan-270) of 270 floor plans with furniture annotations. The method uses a fine-tuned vision-language model and a domain-specific language to generate geometrically valid and functionally plausible furniture layouts that can be rasterized into blueprint-style images.
A scalable framework combines self-distillation and reinforcement learning to transfer task-solving abilities from vision-language models to video diffusion models without requiring labeled task-video data.
OmniGameArena introduces a unified benchmark for evaluating VLM agents in diverse Unreal Engine 5 game environments, featuring an Improvement Dynamics Curve for tracking skill evolution across reflection rounds.
VoLoAgent integrates vision-language models with robot capabilities for open-vocabulary long-horizon manipulation tasks, introducing a physical orchestrator that plans, monitors, and recovers using interruptible tools, and a benchmark called RoboVoLo for evaluation.
A visual guide explaining the full architecture of Gemma 4 12B, covering how it handles text, images, and audio without separate encoder models by removing traditional vision and audio encoders.
This paper introduces Structured Defect Grounding (SDG), a method that models text-to-image defects as structured (location, type, reason, importance) tuples and uses VLMs for detection, along with a 30K-image dataset SDG-30K and a diagnosis-to-alignment framework called BoxFlow-GRPO.
H Company releases Holo3.1, a family of Vision-Language Models (0.8B to 35B) for computer use agents, supporting web, desktop, and mobile automation with native function calling and optimized quantized checkpoints for local deployment.
This paper presents MAOAM, a unified vision-language model framework that enables precise object and material selection through text or click interactions for interactive image editing. It introduces a scalable data generation pipeline and shows emergent improvement when combining text and clicks at inference.
PaddleOCR-VL-1.6 improves document parsing by identifying and refining under-optimized regions via targeted data optimization and progressive post-training, achieving state-of-the-art 96.33% on OmniDocBench v1.6.
MLX-VLM v0.6.0 is released, adding speculative decoding, an agent-ready server compatible with Anthropic's API, new models (DeepSeek V4, ZAYA1-VL, etc.), image generation/editing, and audio input support, enabling local AI agents on Apple devices.
MMG2Skill converts web-based procedural guides into executable skills for agents through closed-loop learning, improving performance across GUI control, gameplay, and card play tasks with macro-average gains of +12.8 to +25.3 percentage points.
PhyDrawGen is a neuro-symbolic pipeline that generates physically accurate diagrams from natural language by combining LLM-based scene understanding with a deterministic constraint solver and a VLM-based verify loop, outperforming existing models on a benchmark of physics problems.
This paper introduces 3DCodeBench, a benchmark for evaluating vision-language models on procedural 3D modeling via code, and 3DCodeArena, a ranking platform based on pairwise human preferences.
This paper introduces the PiSAR benchmark for screen-conditioned action prediction and compares supervised fine-tuned models against frontier zero-shot baselines. Key findings show a fine-tuned Qwen3-VL-8B achieves 0.783 semantic similarity, significantly outperforming Claude Opus 4.7 and GPT-5.5 (0.459 and 0.482), but the same fine-tuning recipe on a larger reasoning-tuned Gemma model yields only 0.441, indicating a model-recipe mismatch.
This paper proposes VFEAgent, a multi-agent system that automates finite element analysis by integrating vision-language models with a verification-first code synthesis framework, enabling end-to-end simulation from images and problem descriptions.
Stable-Layers is a reinforcement learning framework that fine-tunes a pretrained image layer decomposition model using VLM feedback instead of paired supervision, employing Flow-GRPO with LoRA and a two-stage reward calibration pipeline to improve layer quality on the Crello dataset.
This paper introduces VisAnomReasoner, a parameter-efficient vision-language model fine-tuned on a novel benchmark (VisAnomBench) with natural-language rationales, achieving over 21pp improvement in precision and F1 for time-series anomaly detection and strong cross-benchmark generalization.
Proposes MedExpMem, an experience memory framework that enables medical vision-language models to accumulate and retrieve discriminative diagnostic experience from past cases, improving differential diagnosis accuracy by up to 7.0% on a radiology benchmark.
InstructSAM presents a unified framework for multi-instance segmentation using instruction-driven queries that bridge vision-language models and SAM3, achieving strong results across complex benchmarks.