Category
3,573 papers
Across seven benchmarks and four base VLMs, VLM-in-Sandbox achieves the highest sample-weighted average accuracy among Vanilla VLM, Append-only Sandbox, and the proposed method, and is identified as a central abstraction for sandboxed VLM agents.
ConSynth-X, a paired synthetic construction-site image dataset containing 34,199 images derived from 3,109 real-world source scenes, is presented, providing a structured resource for evaluating and improving the robustness of construction vision and vision-language models under challenging field conditions.
This work proposes Targeted Anti-obfuscation with Mechanistic Enforcement (TAME), which uses Sparse Autoencoders (SAEs) to combine behavioral feedback with targeted suppression of template-associated activations during RL, providing a path from behavioral monitoring to representation-level oversight for more auditable...
RoboTalk is introduced, a synthetic data-generation pipeline and dataset of 7,950 multimodal trajectories spanning 53 mobile-manipulation kitchen tasks for training small VLMs to communicate and coordinate and fine-tuning open-source models on this dataset can reach 77% success on novel held-out tasks.
Reach audiences
Advertise in front of researchers, engineers, and readers.
It is argued that office AI is better understood as part of an emerging sociotechnical assemblage than as a tool added to an otherwise stable workplace.
Results indicate that temporal memory is an attack surface that BEV security evaluation has not yet accounted for, and that cross-modal semantic verification is a promising, if not yet field-validated, direction for defending it.
Penicillin allergy is recorded in roughly one in ten hospital patients, yet fewer than one in ten of those records reflects a real, clinically significant allergy. The mismatch matters: patients with a penicillin label receive more second-line and broader-spectrum antibiotics, with measurable downstream costs in infect...
The proposed FastVSearch method bridges efficient data indexing and deep multimodal understanding, enabling practical deployment for large-scale video search applications.
衛生行政報告例(令和6年度)の第5表から、墓地・火葬場・納骨堂の数を経営主体別に数え、I010 で数えた火葬件数と突き合わせた。 墓地は876,042区域、火葬場は3,511施設、納骨堂は13,853施設。47都道府県の合計は4つとも全国の値とぴたり一致する。墓地は火葬場の250倍、恒常的に使用している火葬場の649倍ある。 火葬場3,511のうち、過去1年以内...
🌟 Summary Ultralytics 8.4.158 improves YOLO26 training reliability, model export compatibility, data handling, inference backends, and Ultralytics Platform workflows. 🚀 📊 Key Changes 🔄 Mosaic augmentation now stays closed after OOM recovery Fixed a training issue where automatic batch-size reduction rebuilt the dat...
小規模事業者が置かれている条件を、最低賃金(令和8年度答申)・消費者物価地域差指数(2024年)・1人あたり課税対象所得(令和7年度)・商業地の地価(令和8年 地価公示)の4つで、47都道府県について並べた。 地域差の大きさが、費目によって桁で違う。上と下の比は、消費者物価 1.081倍、最低賃金(現行)1.198倍、1人あたり課税対象所得 1.85倍、商業...
Background: Small Language Models (SLMs) running on local hardware can reduce the privacy, latency, and cost concerns associated with cloud-based Large Language Models (LLMs) in healthcare, but deciding when to use them is typically informal and unsystematic in current practice. The problem is compounded by model turno...
A Blog post by Ai2 on Hugging Face
A Blog post by IBM Research on Hugging Face
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
A Blog post by IBM Granite on Hugging Face
We’re on a journey to advance and democratize artificial intelligence through open source and open science.
Gemini 4 Carbon is our strongest model with state-of-the-art cybersecurity and long horizon capabilities.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
