https://survivingthesingularity.com/signals Algorithmically swept arXiv research ranked by singularity relevance. Not human-curated. en-us Sun, 23 Aug 2026 13:36:12 +0000 https://arxiv.org/abs/2608.07642 https://arxiv.org/abs/2608.07642 Fri, 07 Aug 2026 00:00:00 +0000 Aligning large language models (LLMs) with human values remains a major challenge, especially for trustworthy AI. While existing approaches such as RLHF, CAI, and their variants have achieved promising results, they often rely on a single-agent framework and a unified reward system. This limits thei... [Score: 13.5 | Flagged for: agent, multi-agent, reasoning, alignment, rlhf] Authors: Yuanhong Wu, Djallel Bouneffouf, et al. https://arxiv.org/abs/2608.13900 https://arxiv.org/abs/2608.13900 Fri, 14 Aug 2026 00:00:00 +0000 Large language model (LLM) agents are evolving from conversational assistants into autonomous systems that execute long-horizon tasks through reasoning, tool use, code generation, and workspace manipulation. As agents increasingly operate over persistent environments and multi-step workflows, they f... [Score: 12.0 | Flagged for: agent, agentic, autonomous, tool use, reasoning] Authors: Zhaoyan Sun, Xiaoxiao Wang, et al. https://arxiv.org/abs/2607.27599 https://arxiv.org/abs/2607.27599 Thu, 30 Jul 2026 00:00:00 +0000 Building generalizable agents for diverse applications remains a fundamental challenge. While imitation learning-based policies succeed in specific training environments, they often fail to generalize to novel scenes and tasks. In this work, we propose World Action Planner, a robot planning system t... [Score: 12.0 | Flagged for: agi, world model, agent, planning, reasoning] Authors: Xiangcheng Zhang, Yilun Du https://arxiv.org/abs/2608.06865 https://arxiv.org/abs/2608.06865 Fri, 07 Aug 2026 00:00:00 +0000 The malicious use of generative artificial intelligence to create highly realistic deepfake videos raises serious ethical concerns and poses substantial challenges to AI safety. However, existing deepfake video benchmarks provide limited coverage of recent synthesis methods and generally lack reliab... [Score: 11.0 | Flagged for: agent, multi-agent, reasoning, ai safety, llm] Authors: Xuechao Zou, Shun Zhang, et al. https://arxiv.org/abs/2608.18575 https://arxiv.org/abs/2608.18575 Wed, 19 Aug 2026 00:00:00 +0000 Large language model (LLM)-based multi-agent systems (MAS) often exhibit complex failure modes, which frequently cause agents to produce incorrect outcomes. This motivates the task of Agent Failure Attribution: given a failed multi-agent trajectory, identify the faulty agents and their corresponding... [Score: 10.0 | Flagged for: agent, multi-agent, agentic, reasoning, llm] Authors: Ting-Wei Li, Yuanchen Bei, et al. https://arxiv.org/abs/2608.16747 https://arxiv.org/abs/2608.16747 Mon, 17 Aug 2026 00:00:00 +0000 Many areas of AI research, such as language model interpretability and chain of thought faithfulness, seek to explain model behaviors. But what constitutes a "good" explanation? In this work, we evaluate explanations through the lens of counterfactual simulatability-whether the explanation is useful... [Score: 10.0 | Flagged for: agent, agentic, chain of thought, interpretability, interpret] Authors: Adam Karvonen, Euan Ong, et al. https://arxiv.org/abs/2608.16195 https://arxiv.org/abs/2608.16195 Mon, 17 Aug 2026 00:00:00 +0000 Achieving human-level competitive intelligence and physical agility in humanoid robots remains a profound challenge, particularly in contact-rich and highly dynamic tasks such as boxing. While Multi-Agent Reinforcement Learning offers a principled framework for strategic interaction, its direct appl... [Score: 10.0 | Flagged for: agi, agent, multi-agent, autonomous, humanoid] Authors: Kangning Yin, Kaige Liu, et al. https://arxiv.org/abs/2608.17184 https://arxiv.org/abs/2608.17184 Mon, 17 Aug 2026 00:00:00 +0000 Job Safety Analysis (JSA) and pre-task planning can benefit from prior incident records, yet historical accident data is often stored as unstructured narratives that are difficult to consult at the point of planning. A novel framework centered on large language models (LLMs) for highway construction... [Score: 10.0 | Flagged for: agent, agentic, planning, ai safety, llm] Authors: Mason Smetana, Trevor Neece, et al. https://arxiv.org/abs/2608.06830 https://arxiv.org/abs/2608.06830 Fri, 07 Aug 2026 00:00:00 +0000 Large Language Models (LLMs) are increasingly used as high-level planners in embodied multi-robot systems, enabling robots to interpret natural language instructions and coordinate executable actions. Yet, this growing reliance on LLM planners also raises security concerns. Prior work has focused ma... [Score: 10.0 | Flagged for: agent, multi-agent, interpret, embodied, robot] Authors: Zhen Huang, Zhihuang Liu, et al. https://arxiv.org/abs/2608.06366 https://arxiv.org/abs/2608.06366 Thu, 06 Aug 2026 00:00:00 +0000 Electronic health record (EHR) feature engineering is a major bottleneck in clinical research and AI, accounting for 39-45% of data scientists' workload. This is especially pronounced in heart failure, which affects an estimated 6.7 million U.S. adults and requires integrating fragmented EHR data wi... [Score: 10.0 | Flagged for: agent, multi-agent, reasoning, llm, language model] Authors: Soorya Ram Shimgekar, Michelle Hu, et al. https://arxiv.org/abs/2608.05668 https://arxiv.org/abs/2608.05668 Thu, 06 Aug 2026 00:00:00 +0000 With increasingly diverse and heterogeneous information sources, effectively leveraging multimodal data is becoming pivotal for high-quality financial trading. Although recent advancements in Large Language Model (LLM)-based agents have enabled the ingestion of multimodal inputs, existing methods fa... [Score: 10.0 | Flagged for: agi, agent, agentic, llm, language model] Authors: Changshuo Liu, Yanzheng Jin, et al. https://arxiv.org/abs/2608.03502 https://arxiv.org/abs/2608.03502 Tue, 04 Aug 2026 00:00:00 +0000 Large Language Models (LLMs) have recently shown strong capabilities in reasoning, planning, and tool-use, enabling new forms of autonomous agents. However, LLM-based agents struggle with long-horizon sequential decision tasks that require precise action optimization and environment interaction. Rei... [Score: 10.0 | Flagged for: agent, autonomous, tool-use, planning, reasoning] Authors: Christophe D. Hounwanou, John Emeka Eze, et al. https://arxiv.org/abs/2608.04270 https://arxiv.org/abs/2608.04270 Tue, 04 Aug 2026 00:00:00 +0000 Agentic code generation has shown promise in automating and accelerating software development by utilizing Large Language Models (LLMs) to generate, test, and deploy code. For engineers and scientists, such systems have the potential to accelerate the development of applied and scientific workflows ... [Score: 10.0 | Flagged for: agi, agent, agentic, llm, language model] Authors: Nolan Cutler, Chia-Chen Kuo, et al. https://arxiv.org/abs/2608.00747 https://arxiv.org/abs/2608.00747 Sat, 01 Aug 2026 00:00:00 +0000 Large language models are increasingly integrated into autonomous robotic systems for task planning and control, but this integration exposes them to prompt injection attacks that can lead to unsafe decisions and physical harm. Multi-agent settings increase the risks through cross-agent contaminatio... [Score: 10.0 | Flagged for: agent, multi-agent, autonomous, planning, robot] Authors: Neha Nagaraja, Amisha Bagari, et al. https://arxiv.org/abs/2607.23899 https://arxiv.org/abs/2607.23899 Mon, 27 Jul 2026 00:00:00 +0000 This exploratory study examines whether a large multimodal language model, GPT-5.1, can serve as the high-level controller of a physical mobile robot despite having no prior embodiment, no training in simulated environments, and no exposure to sensorimotor experience. Using only low-resolution first... [Score: 10.0 | Flagged for: world model, reasoning, embodied, robot, language model] Authors: Roberto Spinelli, Thiago C. Martins https://arxiv.org/abs/2607.25140 https://arxiv.org/abs/2607.25140 Mon, 27 Jul 2026 00:00:00 +0000 This paper studies the behavior of language models in a multi-agent crowd simulation, focusing on how affect propagates among agents that perceive and appraise one another. Each agent perceives its neighbors through visual, auditory, and tactile channels, then appraises these perceptions in light of... [Score: 10.0 | Flagged for: agi, agent, multi-agent, llm, language model] Authors: Funda Durupinar https://arxiv.org/abs/2607.22319 https://arxiv.org/abs/2607.22319 Fri, 24 Jul 2026 00:00:00 +0000 Large language models (LLMs) and AI agents have demonstrated strong potential for data integration in zero-shot and few-shot settings. However, they continue to face significant accuracy and cost challenges in enterprise environments due to a persistent knowledge gap. This paper envisions trustworth... [Score: 10.0 | Flagged for: agent, agentic, reasoning, llm, language model] Authors: Chuangtao Ma, Arijit Khan https://arxiv.org/abs/2608.20314 https://arxiv.org/abs/2608.20314 Thu, 20 Aug 2026 00:00:00 +0000 Mid-training is increasingly recognized as a critical stage for shaping the capabilities of large language models. Recent work has shown that targeted mid-training can strengthen reasoning-intensive abilities such as math and science, and can also improve agentic capabilities in software-engineering... [Score: 9.5 | Flagged for: agent, agentic, tool use, tool-use, reasoning] Authors: Fengqing Jiang, Yite Wang, et al. https://arxiv.org/abs/2608.20129 https://arxiv.org/abs/2608.20129 Thu, 20 Aug 2026 00:00:00 +0000 Autonomous vehicles require robust perception and decision-making capabilities to operate in diverse and unseen scenarios. While reinforcement learning and rule-based methods can provide effective control and safety mechanisms, their performance may degrade in situations requiring contextual reasoni... [Score: 9.5 | Flagged for: agent, multi-agent, autonomous, reasoning, llm] Authors: Mehdi Azarafza, Faezeh Pasandideh, et al. https://arxiv.org/abs/2608.16806 https://arxiv.org/abs/2608.16806 Mon, 17 Aug 2026 00:00:00 +0000 Large Language Models (LLMs) have demonstrated capabilities in in-context learning, task decomposition, step-by-step reasoning, and code generation, driving their gradual evolution from text generation models into the core of agents capable of perceiving environments, invoking tools, and executing t... [Score: 9.5 | Flagged for: agent, reasoning, embodied, robot, llm] Authors: Jiawei Liu, Jiacheng Guo, et al. https://arxiv.org/abs/2608.16742 https://arxiv.org/abs/2608.16742 Mon, 17 Aug 2026 00:00:00 +0000 Large Language Models (LLMs) have achieved remarkable progress in code generation, yet ensuring correctness in complex, repository-level tasks remains challenging. Existing approaches often use generated tests as static post-hoc validators, which limits their ability to guide implementation and may ... [Score: 9.5 | Flagged for: agi, agent, reasoning, llm, language model] Authors: Hongyue Yu, Kefan Li, et al. https://arxiv.org/abs/2608.09857 https://arxiv.org/abs/2608.09857 Mon, 10 Aug 2026 00:00:00 +0000 Advances in advanced artificial intelligence tools have sparked research in robot autonomy, but the development of such systems has largely focused on execution rather than verifying the feasibility actions planning models propose. Like general-purpose LLMs, robotics planning models carry risks: bia... [Score: 9.5 | Flagged for: agent, agentic, autonomy, planning, robot] Authors: Rohan Bhagra, Mahantesh Halapannavar, et al. https://arxiv.org/abs/2608.07419 https://arxiv.org/abs/2608.07419 Fri, 07 Aug 2026 00:00:00 +0000 Preference alignment often makes large language models (LLMs) overconfident and poorly calibrated. Traditional post-hoc temperature scaling is inherently domain-dependent: a temperature fitted on one domain does not generalize across domains. This motivates us to modify model parameters during train... [Score: 9.5 | Flagged for: agi, alignment, llm, language model, scaling] Authors: Ruochen Jin, Zhanliang Wang, et al. https://arxiv.org/abs/2608.04309 https://arxiv.org/abs/2608.04309 Wed, 05 Aug 2026 00:00:00 +0000 We present a structured large-language-model (LLM) architecture for zero-shot human--robot coordination in a cooperative construction task with private goal views. Guided by a Dec-POMDP formulation, the architecture decomposes decision-making into (i) action-conditioned Theory-of-Mind (ToM) inferenc... [Score: 9.5 | Flagged for: agent, multi-agent, planning, reasoning, interpret] Authors: Dong Hae Mangalindan, Anand Gokhale, et al. https://arxiv.org/abs/2607.22400 https://arxiv.org/abs/2607.22400 Fri, 24 Jul 2026 00:00:00 +0000 Large Language Models (LLMs) are increasingly deployed as autonomous agents, transitioning from static conversational interfaces to dynamic systems capable of complex reasoning, tool execution, and decision-making. However, the operational reliability of these agentic AI systems is fundamentally cha... [Score: 9.5 | Flagged for: agent, agentic, autonomous, reasoning, llm] Authors: Fin Gentzen, Marla Grunewald, et al. https://arxiv.org/abs/2608.17836 https://arxiv.org/abs/2608.17836 Tue, 18 Aug 2026 00:00:00 +0000 As large language models (LLMs) are granted increasing autonomy, it is essential to investigate methods that can induce unsafe behavior. We propose a novel white-box attack inspired by locate-then-edit approaches from the field of Knowledge Editing. Our choice is motivated by the observation that mo... [Score: 9.0 | Flagged for: agi, autonomy, llm, language model, retrieval] Authors: Roman Maksimov, Vladimir Aletov, et al. https://arxiv.org/abs/2608.16775 https://arxiv.org/abs/2608.16775 Mon, 17 Aug 2026 00:00:00 +0000 Large Language Models (LLMs) are increasingly being deployed in cybersecurity operations to assist cybersecurity analysts with rapid decision-making against emerging threats. However, there is a main criteria that must be met when using LLMs in cybersecurity, that is, trust in the generated outputs.... [Score: 9.0 | Flagged for: agent, agentic, autonomous, llm, language model] Authors: Reza Fayyazi, Michael Zuzak, et al. https://arxiv.org/abs/2608.11679 https://arxiv.org/abs/2608.11679 Wed, 12 Aug 2026 00:00:00 +0000 Digital twins are increasingly used to monitor and simulate the behavior of cyber-physical systems. Even with skilled operators, interpreting anomalies detected within digital twin pipelines is challenging, as the sheer complexity and volume of raw sensor data make thorough analysis difficult. Recen... [Score: 9.0 | Flagged for: agent, agentic, reasoning, interpret, llm] Authors: Touseef Hasan, Mounika Ghanta, et al. https://arxiv.org/abs/2608.05999 https://arxiv.org/abs/2608.05999 Thu, 06 Aug 2026 00:00:00 +0000 Vision-language-action (VLA) models have demonstrated remarkable capabilities in robotic manipulation by leveraging pretrained vision-language models. However, existing post-training methods predominantly optimize VLA models as flat policies, making it difficult to explicitly model task progression ... [Score: 9.0 | Flagged for: agi, agent, embodied, robot, language model] Authors: He Kong, Zengjue Chen, et al. https://arxiv.org/abs/2607.29347 https://arxiv.org/abs/2607.29347 Fri, 31 Jul 2026 00:00:00 +0000 Modern neuroscience relies on integrating multi-scale, multimodal datasets to uncover the neural principles underlying intelligence. However, analytical challenges posed by highly heterogeneous data and fragmented workflows increasingly constrain discoveries. Here we introduce SeekBrain, an autonomo... [Score: 9.0 | Flagged for: agent, multi-agent, agentic, autonomous, planning] Authors: Jiamin Wu, Peishan Xiang, et al.