OmniHallu is a unified framework for detecting hallucinations in multimodal large language models across both comprehension and generation tasks involving image, video, and audio modalities. It introduces OmniHallu-Bench, a 10,000-sample benchmark with claim-level human annotations for six cross-modal tasks (I2T, V2T, A2T, T2I, T2V, T2A). The system uses a multi‑agent architecture that decomposes outputs into atomic claims, verifies them with modality‑specific experts, and aggregates evidence through structured reasoning, while a preference‑optimized verifier reduces expert calls by 66% with minimal performance loss.
By Jianjiang Yang, Peihang Li, Shanqing Xu, Mengchen Qian, Lu Zhang, Meng Luo
The paper presents the KuaiRP series of role‑playing models, detailing a multi‑stage training pipeline that balances deep domain knowledge injection with the preservation of general agent capabilities. The approach includes a standardized character template, a supervised fine‑tuning (SFT) data pipeline, a rule‑based reward function for reinforcement learning, and a novel two‑stage on‑policy distillation (OPD) with cumulative‑divergence decay (CDD) to recover general skills. Experimental results show that the models achieve state‑of‑the‑art role‑playing fidelity in target domains while maintaining low deployment costs.
By Yipeng Wang, Ziwei Zhang, Jiahui Zhang, Qi Gan, Kai Sheng
Deep Research Bench II is a new benchmark designed to evaluate Deep Research Agents (DRAs) by requiring them to produce research reports for 132 grounded tasks across 22 domains. Each report is assessed using 9,430 fine‑grained binary rubrics that cover information recall, analysis, and presentation, all derived from expert‑written investigative articles through a rigorous LLM‑plus‑human pipeline. Evaluation of current state‑of‑the‑art DRAs shows that even the best models satisfy fewer than 50% of these rubrics, highlighting a significant gap between automated agents and human experts.
By Ruizhe Li, Mingxuan Du, Benfeng Xu, Chiwei Zhu, Xiaorui Wang, Zhendong Mao
The paper introduces MERIT, a training‑free agent that uses an online dual‑polarity memory of oracle‑verified corrections and failed attempts to improve LLM agent repair without updating parameters. MERIT retrieves only finalized memories, classifies failures, and employs a hybrid lexical‑dense retriever before the frozen model generates revisions. Experiments with Qwen2.5‑7B‑Instruct show that MERIT raises execution accuracy on the Spider benchmark from 66.34 % to 69.79 % and on BIRD from 47.35 % to 48.44 %, with varying evidence across datasets.
By Khang Nhat Hoang Vo, Tam Minh Chu, Anh Trac Duc Dinh, Thuyen Vinh Ha Bui, Tho Quan
The paper introduces a new framework for learning continuous-time diffeomorphic image registration by modeling a non-autonomous ODE as a two-parameter flow map. By enforcing cocycle consistency, the method learns flow maps without time discretization or velocity integration during training, enabling efficient inference with few compositions. Experiments on nine datasets show consistent alignment improvements, including a 2.1% Dice gain on brain MRI, 12% TRE reduction on lung CT, and 2.6% Dice improvement on cardiac MRI and ultrasound.
By Mohammadjavad Matinkia, Nilanjan Ray
The paper presents a decision‑focused active learning framework for optimizing scale‑aware recovery of critical materials, using data from the Pacific Northwest National Laboratory’s CICERO workflow on autonomous selective precipitation. By adaptively selecting experiments based on prior results, the method achieves the best recorded enrichment of recycled neodymium‑iron‑boron magnets with roughly half the number of experiments compared to non‑adaptive approaches, and demonstrates similar efficiency for samarium‑cobalt magnets. The authors propose a Bayesian risk‑based batch selection strategy and outline a prospective test plan with standardized logging and economic validation.
By Niranjan Srinivas, Debajyoti Ray, Elias Nakouzi
The paper introduces the Belief-State Engine (BSE), an inference module that supplies a large language model (LLM) with a Bayesian posterior over hidden states in a partially observable Markov decision process (POMDP). By keeping the raw action‑observation log hidden from the LLM, the BSE ensures the agent behaves as a sound Markov policy on the belief MDP, thereby inheriting classical POMDP optimality guarantees. Experiments on the Tiger POMDP and a red‑team attack‑graph task show that BSE‑augmented agents outperform six baselines in task return, belief calibration, and decision consistency.
By Arnab Chattopadhayay, Debdipta Halder
The paper presents an algorithm that lets a learning agent ask for help from a mentor and transfer knowledge between similar states, enabling safe and effective learning in Markov decision processes with irreversible dynamics and infinite state spaces. It proves that both regret and the number of mentor queries grow sublinearly over time, using a sequence of three reductions to achieve a general result. The work claims to be the first formal proof that an agent can achieve high reward while becoming self‑sufficient in an unknown, unbounded, high‑stakes environment without resets.
By Benjamin Plaut, Juan Li\'evano-Karim, Hanlin Zhu, Stuart Russell
The paper proposes a shift from AI model storage to AI model management, introducing the concept of "learnware"—a combination of a model and its specification. Learnware specifications are generated without exposing developers’ training data, enabling models from different sources to be identified, reused, and assembled for new tasks. The Learnware Dock System (LDS) offers a framework for managing these learnwares and facilitates collaboration among independently developed models through a shared specification protocol.
By Zhi-Hua Zhou
OpenDiscoveryTrace is a public dataset of 558 complete AI scientific agent trajectories that records the reasoning process—thoughts, tool calls, observations, errors, revision triggers, and confidence—across 124 scientific tasks in drug discovery, materials science, genomics, and literature analysis. The dataset includes seven models (three frontier models and four open‑weight models) and 60 live‑retrieval variants, providing a balanced view of performance and error patterns. Pilot analysis shows that process traces reveal behavioral differences invisible to output‑only evaluation, such as differing error rates and types among frontier models.
By Aayam Bansal, Keertan Balaji
The paper presents a probability‑wave framework for modeling the collective behavior of adaptive agents, deriving testable eigenmodes via a generalized behavioral intelligence (GBI) nonlocal probability‑wave equation. Empirical analysis of Chinese intraday stock market data shows that adaptive entangled game modes explain 89% of observed decision patterns, far exceeding predictions from neoclassical finance and supporting the Liu‑Chen‑Ao hypothesis of nonlocal entangled nerve fibers. The authors argue that incorporating these adaptive entangled game modules into AGI architectures can overcome limitations of conventional ANN‑based AI and enable more compact, efficient, and robust human‑like processing units for embodied intelligence and robotics.
By Haochen Li, Xinshuai Guo, Jingdong Ouyang, Wei Zhang, Leilei Shi
An autonomous GeoAI agent for Arctic eco‑navigation is presented, integrating operational, physical, ecological, and community criteria into a unified routing framework. The system employs multiple specialized agents for data acquisition, multi‑objective route generation, and skyline‑based decision support, explicitly accounting for exposure to sensitive ecological areas such as Essential Fish Habitat and seal critical habitat. By keeping value judgments under human control, the framework aims to produce safer, more transparent, and socially responsible Arctic navigation routes.
By Samira Alkaee Taleghan, Younghyun Koo, Farnoush Banaei-Kashani
The paper introduces Multi-Agent Agentic Graph Learning (MAAGL), a framework that partitions a graph into communities and assigns a dedicated agent to each community for specialized reasoning. MAAGL addresses two key challenges in existing agentic graph learning: it preserves permutation invariance by summarizing structural evidence with a dynamic structural signature, and it controls context size by filtering semantic evidence to the top‑k relevant nodes. Experiments on four benchmark datasets demonstrate that MAAGL outperforms state‑of‑the‑art agentic graph learning methods.
By Liang Qu, Jianxin Li, Hua Wang
The paper introduces a black-box framework for evaluating agentic AI systems, focusing on multi-step vulnerabilities that standard single-turn tests miss. It presents a seven-domain taxonomy linking observable behaviors to risk categories, an automated SAGE-RT red-teaming process generating 120 adversarial scenarios per domain, and a human-validated evaluation using LLM judges. Empirical tests on CrewAI and AutoGen agents show significant governance, privacy, and behavior risks, demonstrating the framework’s ability to uncover critical architectural weaknesses without privileged access.
By Divyanshu Kumar, Nitin Aravind Birur, Tanay Baswa, Sahil Agarwal, Prashanth Harshangi
The paper introduces SmartWeatherAgent, a three‑stage architecture that combines intent recognition, hazard prediction, and reasoning‑enhanced generation to improve tourism meteorological services. It fuses rule‑based methods with large language models and a LightGBM model enriched with highland‑specific features, achieving an F1‑Macro score of 0.605 and 1.60 ms latency on high‑wind, precipitation, and low‑temperature events. A 12‑round micro‑step prompt self‑optimization loop raises the composite warning quality score from 4.2 to 8.9, with notable gains in data source citation, physical mechanism explanation, and scientific rigor through explicit uncertainty statements.
By Shuai Yan, Yang Xu, Shan He
HiRAD is a hierarchical reinforcement learning framework designed for continuous-space routing of large-scale AGV fleets, offering real-time guarantees. It introduces a step-level spatiotemporal representation, separates heading selection from velocity control to shrink the action space, and employs an asynchronous event-driven decision pipeline that reduces inference complexity from O(n²) to O(n) and cuts per-step latency by up to 71%. Experiments on random graphs and two warehouse maps show that HiRAD decreases makespan by 45% to 63% and shortens overall runtime.
By Yunjie Huang, Ruizhong Wu, Mengxuan Zhang, Frodo Kin Sun Chan, Yan Nei Law, Lei Li
The paper introduces Agentic Just-In-Time Software Construction (A-JIT), a paradigm that replaces static software binaries with dynamic systems capable of continuous evolution. In A-JIT, an application consists of code, a runtime harness, and an embedded AI agent that observes usage and execution traces to specialize software logic, workflows, and tool interfaces for each user. This approach enables applications to dynamically generate missing implementations, create new capabilities on the fly, and adapt continuously to end‑user behavior, thereby supporting trace‑driven human‑AI co‑construction and opening a new design space for adaptive, self‑evolving software.
By Mark Marron, Earl T. Barr
Grounded Continuation introduces a runtime verifier that classifies each utterance in an LLM conversation into one of eight epistemic operations and uses a symbolic engine to maintain a dependency map of claims and their supports. The verifier checks whether a new continuation is grounded by walking this map, a linear-time process that requires no additional LLM calls. On benchmarks such as ReviseQA and MemoryAgentBench, the verifier improves single-hop accuracy for several QA models, even enabling a 7B model to outperform GPT‑4o when guided by the verifier.
By Qisong He, Jinwei Hu, Xinmiao Huang, Changshun Wu, Yi Dong, Xiaowei Huang
The paper presents a method where self‑evolving scientific agents design explicit, neural‑network‑free white‑box controllers for fluid dynamics tasks. By iteratively interpreting simulation data, the agents accumulate control knowledge and refine controller code, ultimately achieving robust control of an underactuated two‑joint swimmer in unsteady flows. The resulting controllers generalize across varying target positions, wake geometries, cylinder counts, and inflow speeds, and 2D control priors successfully transfer to accelerate 3D adaptation.
By Boai Sun, Wenjin Guo, Zongmin Yu, Liu Yang
Spec‑Harness evaluates how well large language models (LLMs) synthesize Java Modeling Language (JML) specifications by measuring behavioral adequacy across precondition and postcondition correctness and completeness. The study shows that while prompt optimization can raise verifier pass rates, many accepted specifications remain behaviorally weak, either over‑ or under‑constraining inputs and outputs. Spec‑Harness also serves as a feedback mechanism that improves the quality of specifications generated by general‑purpose coding agents and a specialized JML agent.
By Md Rakib Hossain Misu, Iris Ma, Cristina V. Lopes