The use of large language models (LLMs) to simulate human behavior has gained significant attention, particularly through personas that approximate individual characteristics. Persona-based simulations hold promise for transforming disciplines that rely on population-level feedback, including social science, economic analysis, marketing research, and business operations. Traditional methods to collect realistic persona data face significant challenges: they are prohibitively expensive and logistically challenging due to privacy constraints, and often fail to capture multi-dimensional attributes, particularly subjective qualities. Consequently, synthetic persona generation with LLMs offers a scalable, cost-effective alternative. However, current approaches rely on ad hoc and heuristic generation techniques that do not guarantee methodological rigor or simulation precision, resulting in systematic biases in downstream tasks. Through extensive large-scale experiments including presidential election forecasts and general opinion surveys of the U.S. population, we reveal that these biases can lead to significant deviations from real-world outcomes. Based on the experimental results, this position paper argues that a rigorous and systematic science of persona generation is needed to ensure the reliability of LLM-driven simulations of human behavior. We call for not only methodological innovations and empirical foundations but also interdisciplinary organizational and institutional support for the development of this field. To support further research and development in this area, we have opensourced approximately one million generated personas, available for public access and analysis at Tianyi-Lab/Personas.

Large language model (LLM) agents have demonstrated remarkable capabilities in complex reasoning and decision-making by leveraging external tools. However, this tool-centric paradigm introduces a previously underexplored attack surface, where adversaries can manipulate tool metadata--such as names, descriptions, and parameter schemas--to influence agent behavior. We identify this as a new and stealthy threat surface that allows malicious tools to be preferentially selected by LLM agents, without requiring prompt injection or access to model internals. To demonstrate and exploit this vulnerability, we propose the Attractive Metadata Attack (AMA), a black-box in-context learning framework that generates highly attractive but syntactically and semantically valid tool metadata through iterative optimization. The proposed attack integrates seamlessly into standard tool ecosystems and requires no modification to the agent's execution framework.

Unlocking spatial reasoning in Multimodal Large Language Models (MLLMs) is crucial for enabling intelligent interaction with 3D environments. While prior ef ask: forts can often MLLMs rely on reason explicit about 3D 3D inputs space or specialized using only model structur architectures, ed 2D represenwe tations derived from perception? We introduce Struct2D, a perception-guided prompting marks and object-centric framework that metadata, combines optionally bird's-eye-vie incorporating w (BEV) egocentric images with keyframes object when needed. Using Struct2D, we conduct an in-depth zero-shot analysis of closed-source spatial reasoning MLLMs abilities (e.g.

Filtering Step Remaining Instances Raw images collected 7.5Mimages After deduplication and car detection 4.9M images Initial car bounding boxes 13.22M boxes After score/size thresholding 6.97M boxes After Qwen2.5-7B

Below we describe a few other prevalent multi-label datasets and explain how the ML48S differs800 from them, hence they were excluded from comparison in this paper.801 PASCALVOC [11] was created for object detection and classification, covering 20 basic-level802 classes across 4,574 images, with most images containing a single prominent object. This dataset is803 much smaller than ML48S and also contains much fewer classes which are all coarse-grained.804 VG500 is a modification of the Visual Genome dataset [19], a dataset focused on dense annotations805 linking images to respective captions. This dataset is not intended to be bounded by categories806 but has open-vocabulary annotations.

A.1 Dataset Examples450 In this section of the appendix, we present a detailed overview of several representative tasks from451 each category included in REASONINGGYM. For each task, we describe its structure, complexity452 parameters, and provide examples.453 A.1.1 complex_arithmetic(Algebra)454 Find the solution of an arithmetic operation involving complex numbers.455 The spiral order is clockwise, starting from the top-left corner. Predict the corresponding output grid by applying the rule you found.

We introduce PurpCode, the first post-training recipe for training safe code reasoning models towards generating secure code and defending against malicious cyberactivities. PurpCode trains a reasoning model in two stages: (i) Rule Learning, which explicitly teaches the model to reference cybersafety rules to generate vulnerabilityfree code and to avoid facilitating malicious cyberactivities; and (ii) Reinforcement Learning, which optimizes model safety and preserves model utility through diverse, multi-objective reward mechanisms. To empower the training pipelines with comprehensive cybersafety data, we conduct internal red-teaming to synthesize comprehensive and high-coverage prompts based on real-world tasks for inducing unsafe cyberactivities in the model. Based on PurpCode, we develop a reasoning-based coding model, namely PurpCode-32B, which demonstrates state-of-the-art cybersafety, outperforming various frontier models. Moreover, our alignment method decreases the model overrefusal rates in both general and cybersafety-specific scenarios, while preserving model utility in both code generation and common security knowledge.

In this section, we introduce the construction pipeline for generating MVU-Eval QA pairs based on2 each data source.3 These questions include: (1) Object Recognition, (2)8 Spatial Understanding, (3) Counting, (4) Knowledge-intensive Reasoning, and (5) Temporal9 Reasoning. These generated questions, answers, and candidate choices are manually checked by10 humans. Pipelines for constructing video pairs are slightly different across datasets.11 By default, 2-6 videos are randomly sampled, regardless of their labels.

Although RAW images offer advantages over sRGB by avoiding ISP-induced distortion and preserving more information in low-light conditions, their widespread use is limited due to high storage costs, transmission burdens, and the need for significant architectural changes for downstream tasks. To address the issues, this paper explores a new raw-based machine vision paradigm, termed Compact RAW Metadata-guided Image Refinement (CRM-IR). In particular, we propose a Machine Vision-oriented Image Refinement (MV-IR) module that refines sRGB images to better suit machine vision preferences, guided by learned raw metadata. In detail, we propose a Cross-Modal Contextual Entropy (CMCE) network for raw metadata extraction and compression. It builds upon the latent representation and entropy modeling framework of learned image compression methods, and uniquely exploits the contextual correspondence between raw images and their sRGB counterparts to achieve more efficient and compact metadata representation. Additionally, we integrate priors derived from the ISP pipeline to simplify the refinement process, enabling a more efficient design. Such a design allows the CRM-IR to focus on extracting the most essential metadata from raw images to support downstream machine vision tasks, while remaining plug-and-play and fully compatible with existing imaging pipelines, without any changes to model architectures or ISP modules. We implement our CRM-IR scheme on various object detection networks, and extensive experiments under low-light conditions demonstrate that it can significantly improve performance with an additional bitrate cost of less than 10 3 bits per pixel.

Estimating agent pose and 3D scene structure from multi-camera rigs is a central task in embodied AI applications such as autonomous driving. Recent learned approaches such as DUSt3R have shown impressive performance in multiview settings. However, these models treat images as unstructured collections, limiting effectiveness in scenarios where frames are captured from synchronized rigs with known or inferable structure. To this end, we introduce Rig3R, a generalization of prior multiview reconstruction models that incorporates rig structure when available, and learns to infer it when not. Rig3R conditions on optional rig metadata including camera IDs, timestamp, and rig calibrations to develop a rig-aware latent space that remains robust to missing information.