Wyvern is proposed, a multi-agent framework for the automated generation of grounded, multimodal technical reports, allowing for the generation of multimodal outputs, integrating images, tables, and text with supporting references in a unified report.
Abstract
In the current artificial intelligence-driven innovation era, the pace of knowledge growth is accelerating, and is hard to keep up with. While generative models are increasingly used to synthesize content, they often lack in information grounding. To address these peculiarities of our time, we propose Wyvern, a multi-agent framework for the automated generation of grounded, multimodal technical reports. Wyvern allows for the generation of multimodal outputs, integrating images, tables, and text with supporting references in a unified report. Additionally, a particular focus is placed on the grounding of the content, with the implementation of a claims auto-revision stage. We conduct a human evaluation study to assess the quality of our proposed framework. The results show that the figures'informativeness is perceived as superior to that of a recent baseline in 87% of cases. Furthermore, Wyvern's reports are rated as more useful than those produced by three alternative methods in 63% to 100% of instances. We also carry out automatic evaluations showing that Wyvern gains up to 2.3$\times$ in citation recall and 1.6$\times$ in citation precision with respect to the baselines.
Despite the growing use of LLM-powered agents to solve multi-step tasks in complex environments, existing benchmarks rarely test state preservation, performance across languages, and application to realistic, grounded scenarios. To address these concerns, we present WorldBench: a comprehensive, multilingual benchmark of genuine, persona-grounded everyday workflows, where agents can act in a sandbox via structured actions. WorldBench comprises 1,600 tasks across seven languages and eight cultures, filtered and refined through feedback from human annotators with language- and culture-specific expertise. For evaluation, we extend metrics from previous works and introduce Constrained Task Success (CTS), which combines natural language instructions and testbeds to score task completion, minimal modification, and other complementary metrics through deterministic and LLM-as-a-Judge evaluations. Our experiments show that frontier models reach only 49.2% CTS, with all models demonstrating large gaps between correctness and environment preservation. We thereby show that current agents remain brittle in multilingual, agentic scenarios, especially for long-horizon tasks and under state-preservation constraints
Leonardo Ranaldi, Sherrie Shen, Jushi Kai et al.· 0 citations
Thanks to the development of basic models and the high quality of the data, the emergence of AI-generated content has
accelerated. Despite its incredible success, there are still challenges that are yet to be addressed in AI content generation, such as
the processing of long-trail information, maintaining up-to-date knowledge, addressing high inference and training expenses,
and addressing data leakage. The shift to address those challenges has been called Retrieval Augmented Generation (RAG).
RAG has brought the process of gathering information, which improves the process of data generation by recovering relevant
information from available data sources, resulting in robustness and accuracy. RAG has become the foundation of Natural
Language Processing (NLP) to effectively fill the gap between factual accuracy of knowledge and fluency of “Large Language
Models (LLMs)”. This study traces the root of RAG from its beginnings as a framework for knowledge-based tasks to its current
state as an agentic, modular and complex runtime of knowledge. This study is an in-depth analysis of the evolution of Naïve
RAG to Modular and Advanced RAG models, and the introduction of new innovations, such as self-reflection, dense vector
recovery, and the use of different models. They are then examined to provide detailed feedback on how to make RAG truly
dynamic and usable as a verifier when applying them to organizations.
Mallikarjunarao Sunke, S. Gudi, Sriharsha Gudi· International Journal for Re...· 0 citations
The rapid evolution of Generative AI (GenAI) has transformed the ways in which knowledge is created, shared, interpreted, and applied in organization and educational contexts. While earlier studies have often focused on the technical capabilities of GenAI or on the detection of synthetic content, less attention has been given to how these tools influence knowledge management processes at the individual and group levels. This study addresses that gap by examining how GenAI affects the production, transfer, validation, and use of knowledge among individuals and within collaborative settings. The aim of this study is to explore the current state of 'AI vs AI' by generating a dataset comprising texts, images, audio and video, utilizing a set of freely available AI tools. This pilot illustrative study was based on a two-stage strategy: in the first stage synthetic media was generated with different GenAI tools to mimic human creativity. A second stage, which involved a rigorous evaluation of AI-detection tools for each modality separately. Texts were created in three different GenAI tools and Images were prompted to mimic Renaissance paintings by Michelangelo and Raphael. Audio was generated with the purpose of mimicking interviews with the Renaissance painters. Videos, finally, were created to be deceptive deepfakes, placing persons in environments and situations where they never have been in real life. Findings provide a comparative analysis of tools' accuracy and false-positive rates that could act as valuable guidelines in the increasing flood of AI-generated content. This research contributes to the growing discussion on how organizations and institutions can manage the opportunities and risks associated with GenAI in knowledge-intensive environment. The study highlights the need for critical AI literacy, transparent knowledge practices, and governance mechanisms that ensure the responsible integration of GenAI into individual and collaborative knowledge work.
Anastasiia Iufereva, Peter Mozelius· European Conference on Knowl...· 0 citations
A taxonomy of core techniques, a layered system architecture and architectural paradigms, reviews representative implementations and applications, and highlights open challenges and future directions are provided.
Jinyan Cai· International journal of eng...· 0 citations
This tutorial provides an in-depth treatment of modern RAG based on AI-facilitated systematic analysis of ~2000 recent papers (2020--2026) and traces the RAG pipeline from its modular foundations through graph-enhanced reasoning to the latest RL-driven agentic architectures, covering each stage.
X. Dong, Sanat Sharma, Kai Sun et al.· Proceedings of the 32nd ACM...· 0 citations
Creative AI is moving from single-step asset generation toward long-horizon multimodal production. Although recent generative models can synthesize high-quality images, videos, audio clips, UI elements, storyboards, slides, and other creative assets, real-world creative work requires more than isolated prompt-output interactions. It involves references, drafts, alternatives, edits, failed attempts, version relations, tool actions, evaluation signals, and human feedback, which together form an evolving project state. Existing prompt-based, chat-based, and node-based generation systems only partially support this state, as they often discard intermediate context, rely on linear conversations, or require manually specified workflows. Recent commercial systems indicate a shift toward agent-assisted creative production, but their closed architectures make it difficult to study how agents represent context, choose tools, revise artifacts, recover from failures, and maintain consistency over time. To address this gap, we introduce JarvisHub, a canvas-native creative agent harness for long-horizon multimodal creation. JarvisHub treats an editable canvas as the user workspace, the agent's external memory, action space, and shared project state, representing multimodal artifacts, dependencies, versions, and feedback as typed canvas nodes and links. Through a three-layer architecture of canvas state, protocol bridge, and agent runtime, JarvisHub enables agents to act within an inspectable and editable creative state. This design moves creative agents beyond isolated tool use toward sustained, human-steerable creative automation, where agents can progressively plan, generate, revise, and organize multimodal projects while users remain able to inspect, guide, and intervene throughout the process.
