Related papers: SCHEMA for Gemini 3 Pro Image: A Structured Method…
Graphical User Interface (GUI) grounding - the task of mapping natural language instructions to screen coordinates - is essential for autonomous agents and accessibility technologies. Existing systems rely on monolithic models or one-shot…
We propose EMMA, an efficient and unified architecture for multimodal understanding, generation and editing. Specifically, EMMA primarily consists of 1) An efficient autoencoder with a 32x compression ratio, which significantly reduces the…
As autonomous agents become more capable of performing real-world tasks, distinguishing scheming behavior from benign task pursuit may become a central AI control problem. Existing monitors often rely on chain-of-thought access or internal…
Text-based prompting remains the predominant interaction paradigm in generative AI, yet it often introduces friction for novice users such as small business owners (SBOs), who struggle to articulate creative goals in domain-specific…
Foundation models such as the recently introduced Segment Anything Model (SAM) have achieved remarkable results in image segmentation tasks. However, these models typically require user interaction through handcrafted prompts such as…
This paper presents SPIE: a novel approach for semantic and structural post-training of instruction-based image editing diffusion models, addressing key challenges in alignment with user prompts and consistency with input images. We…
Benchmarks are the primary tool for assessing progress in artificial intelligence (AI), yet current practice evaluates models on isolated test suites and provides little guidance for reasoning about generality or autonomous…
Humans can intuitively compose and arrange scenes in the 3D space for photography. However, can advanced AI image generators plan scenes with similar 3D spatial awareness when creating images from text or image prompts? We present GenSpace,…
This paper introduces MeLA, a Metacognitive LLM-Driven Architecture that presents a new paradigm for Automatic Heuristic Design (AHD). Traditional evolutionary methods operate directly on heuristic code; in contrast, MeLA evolves the…
With the proposal of the Segment Anything Model (SAM), fine-tuning SAM for medical image segmentation (MIS) has become popular. However, due to the large size of the SAM model and the significant domain gap between natural and medical…
Labeling data (e.g., labeling the people, objects, actions and scene in images) comprehensively and efficiently is a widely needed but challenging task. Numerous models were proposed to label various data and many approaches were designed…
Traditional ML models utilize controlled approximations during high loads, employing faster, but less accurate models in a process called accuracy scaling. However, this method is less effective for generative text-to-image models due to…
Due to their architecture and vast pre-training data, large language models (LLMs) demonstrate strong text classification performance. However, LLM output - here, the category assigned to a text - depends heavily on the wording of the…
As AI-generated images become increasingly photorealistic, distinguishing them from natural images poses a growing challenge. This paper presents a robust detection framework that leverages multiple uncertainty measures to decide whether to…
What symbolic format should an LLM emit for reliable 3D scene editing from natural language, and does algebraic structure help beyond compact syntax? We evaluate Conformal Geometric Algebra (CGA) as a compact symbolic interface against a…
Unified multimodal models target joint understanding, reasoning, and generation, but current image editing benchmarks are largely confined to natural images and shallow commonsense reasoning, offering limited assessment of this capability…
The proliferation of large language models (LLMs) in educational settings has paradoxically undermined the cognitive processes they purport to support. Students increasingly outsource critical thinking to AI assistants that generate…
We introduce scheming honeypot evaluations, a framework for testing whether models will pursue instrumental goals if given the opportunity. Our scheming honeypot evaluations take the form of coding tasks in Google's alignment research…
Prompt injection defenses are often framed as semantic understanding problems and delegated to increasingly large neural detectors. For the first screening layer, however, the requirements are different: the detector runs on every request…
Text-to-image models are enabling efficient design space exploration, rapidly generating images from text prompts. However, many generative AI tools are imperfect for product design applications as they are not built for the goals and…