As large-scale multi-agent systems evolve, the communication protocol layer has become a critical yet under-evaluated factor shaping performance and reliability. Despite the existence of diverse protocols (A2A, ACP, ANP, Agora, etc.), the selection of them is often intuition-driven and lacks standardized guidance. We introduce ProtocolBench, a benchmark that systematically compares agent protocols along four measurable axes: task success, end-to-end latency, message or byte overhead, and robustness under failures. On ProtocolBench, the choice of protocol significantly influences system behavior. In the Streaming Queue scenario, overall completion time varies by up to 36.5% across protocols, and mean end-to-end latency differs by 3.48 s. Under Fail-Storm Recovery, resilience also differs consistently across protocols. Beyond evaluation, we present ProtocolRouter, a learnable protocol router that selects per-scenario (or per-module) protocols from requirement and runtime signals. ProtocolRouter reduces Fail-Storm recovery time by up to 18.1% versus the best single-protocol baseline, and achieves scenario-specific gains such as higher success in GAIA. We also release ProtocolRouterBench to standardize protocol evaluation and improve reliability at scale.

The rapid evolution of agentic AI marks a new phase in artificial intelligence, where Large Language Models (LLMs) no longer merely respond but act, reason, and adapt. This survey traces the paradigm shift in building agentic AI: from Pipeline-based systems, where planning, tool use, and memory are orchestrated by external logic, to the emerging Model-native paradigm, where these capabilities are internalized within the model's parameters. We first position Reinforcement Learning (RL) as the algorithmic engine enabling this paradigm shift. By reframing learning from imitating static data to outcome-driven exploration, RL underpins a unified solution of LLM + RL + Task across language, vision and embodied domains. Building on this, the survey systematically reviews how each capability -- Planning, Tool use, and Memory -- has evolved from externally scripted modules to end-to-end learned behaviors. Furthermore, it examines how this paradigm shift has reshaped major agent applications, specifically the Deep Research agent emphasizing long-horizon reasoning and the GUI agent emphasizing embodied interaction. We conclude by discussing the continued internalization of agentic capabilities like Multi-agent collaboration and Reflection, alongside the evolving roles of the system and model layers in future agentic AI. Together, these developments outline a coherent trajectory toward model-native agentic AI as an integrated learning and interaction framework, marking the transition from constructing systems that apply intelligence to developing models that grow intelligence through experience.

Integration of Large Language Models with search/retrieval engines has become ubiquitous, yet these systems harbor a critical vulnerability that undermines their reliability. We present the first systematic investigation of "chameleon behavior" in LLMs: their alarming tendency to shift stances when presented with contradictory questions in multi-turn conversations (especially in search-enabled LLMs). Through our novel Chameleon Benchmark Dataset, comprising 17,770 carefully crafted question-answer pairs across 1,180 multi-turn conversations spanning 12 controversial domains, we expose fundamental flaws in state-of-the-art systems. We introduce two theoretically grounded metrics: the Chameleon Score (0-1) that quantifies stance instability, and Source Re-use Rate (0-1) that measures knowledge diversity. Our rigorous evaluation of Llama-4-Maverick, GPT-4o-mini, and Gemini-2.5-Flash reveals consistent failures: all models exhibit severe chameleon behavior (scores 0.391-0.511), with GPT-4o-mini showing the worst performance. Crucially, small across-temperature variance (less than 0.004) suggests the effect is not a sampling artifact. Our analysis uncovers the mechanism: strong correlations between source re-use rate and confidence (r=0.627) and stance changes (r=0.429) are statistically significant (p less than 0.05), indicating that limited knowledge diversity makes models pathologically deferential to query framing. These findings highlight the need for comprehensive consistency evaluation before deploying LLMs in healthcare, legal, and financial systems where maintaining coherent positions across interactions is critical for reliable decision support.

As Large Language Model (LLM) agents increasingly operate in complex environments with real-world consequences, their safety becomes critical. While uncertainty quantification is well-studied for single-turn tasks, multi-turn agentic scenarios with real-world tool access present unique challenges where uncertainties and ambiguities compound, leading to severe or catastrophic risks beyond traditional text generation failures. We propose using "quitting" as a simple yet effective behavioral mechanism for LLM agents to recognize and withdraw from situations where they lack confidence. Leveraging the ToolEmu framework, we conduct a systematic evaluation of quitting behavior across 12 state-of-the-art LLMs. Our results demonstrate a highly favorable safety-helpfulness trade-off: agents prompted to quit with explicit instructions improve safety by an average of +0.39 on a 0-3 scale across all models (+0.64 for proprietary models), while maintaining a negligible average decrease of -0.03 in helpfulness. Our analysis demonstrates that simply adding explicit quit instructions proves to be a highly effective safety mechanism that can immediately be deployed in existing agent systems, and establishes quitting as an effective first-line defense mechanism for autonomous agents in high-stakes applications.

Corporate responsibility turns on notions of corporate \textit{mens rea}, traditionally imputed from human agents. Yet these assumptions are under challenge as generative AI increasingly mediates enterprise decision-making. Building on the theory of extended cognition, we argue that in response corporate knowledge may be redefined as a dynamic capability, measurable by the efficiency of its information-access procedures and the validated reliability of their outputs. We develop a formal model that captures epistemic states of corporations deploying sophisticated AI or information systems, introducing a continuous organisational knowledge metric $S_S(φ)$ which integrates a pipeline's computational cost and its statistically validated error rate. We derive a thresholded knowledge predicate $\mathsf{K}_S$ to impute knowledge and a firm-wide epistemic capacity index $\mathcal{K}_{S,t}$ to measure overall capability. We then operationally map these quantitative metrics onto the legal standards of actual knowledge, constructive knowledge, wilful blindness, and recklessness. Our work provides a pathway towards creating measurable and justiciable audit artefacts, that render the corporate mind tractable and accountable in the algorithmic age.

Supply chain attacks significantly threaten software security with malicious code injections within legitimate projects. Such attacks are very rare but may have a devastating impact. Detecting spurious code injections using automated tools is further complicated as it often requires deciphering the intention of both the inserted code and its context. In this study, we propose an unsupervised approach for highlighting spurious code injections by quantifying cohesion disruptions in the source code. Using a name-prediction-based cohesion (NPC) metric, we analyze how function cohesion changes when malicious code is introduced compared to natural cohesion fluctuations. An analysis of 54,707 functions over 369 open-source C++ repositories reveals that code injection reduces cohesion and shifts naming patterns toward shorter, less descriptive names compared to genuine function updates. Considering the sporadic nature of real supply-chain attacks, we evaluate the proposed method with extreme test-set imbalance and show that monitoring high-cohesion functions with NPC can effectively detect functions with injected code, achieving a Precision@100 of 36.41% at a 1:1,000 ratio and 12.47% at 1:10,000. These results suggest that automated cohesion measurements, in general, and name-prediction-based cohesion, in particular, may help identify supply chain attacks, improving source code integrity.

This paper establishes analytical stability criteria for robot-mediated human-human (dyadic) interaction systems, focusing on haptic communication under network-induced time delays. Through frequency-domain analysis supported by numerical simulations, we identify both delay-independent and delay-dependent stability criteria. The delay-independent criterion guarantees stability irrespective of the delay, whereas the delay-dependent criterion is characterised by a maximum tolerable delay before instability occurs. The criteria demonstrate dependence on controller and robot dynamic parameters, where increasing stiffness reduces the maximum tolerable delay in a non-linear manner, thereby heightening system vulnerability. The proposed criteria can be generalised to a wide range of robot-mediated interactions and serve as design guidelines for stable remote dyadic systems. Experiments with robots performing human-like movements further illustrate the correlation between stability and motor performance. The findings of this paper suggest the prerequisites for effective delay-compensation strategies.

The emergence of large language models (LLMs) has opened new opportunities for creating dynamic non-player characters (NPCs) in gaming environments, enabling both functional task execution and persona-consistent dialogue generation. In this paper, we (Tu_Character_lab) report our participation in the Commonsense Persona-Grounded Dialogue Challenge (CPDC) 2025 Round 2, which evaluates agents across three tracks: task-oriented dialogue, context-aware dialogue, and their integration. Our approach combines two complementary strategies: (i) lightweight prompting techniques in the API track, including a Deflanderization prompting method to suppress excessive role-play and improve task fidelity, and (ii) fine-tuned large models in the GPU track, leveraging Qwen3-14B with supervisedfinetuning (SFT) and Low-Rank Adaptation(LoRA). Our best submissions ranked 2nd on Task 1, 2nd on Task 3 (API track), and 4th on Task 3 (GPU track).

In many domains, including online education, healthcare, security, and human-computer interaction, facial emotion recognition (FER) is essential. Real-world FER is still difficult despite its significance because of some factors such as variable head positions, occlusions, illumination shifts, and demographic diversity. Engagement detection, which is essential for applications like virtual learning and customer services, is frequently challenging due to FER limitations by many current models. In this article, we propose ExpressNet-MoE, a novel hybrid deep learning model that blends both Convolution Neural Networks (CNNs) and Mixture of Experts (MoE) framework, to overcome the difficulties. Our model dynamically chooses the most pertinent expert networks, thus it aids in the generalization and providing flexibility to model across a wide variety of datasets. Our model improves on the accuracy of emotion recognition by utilizing multi-scale feature extraction to collect both global and local facial features. ExpressNet-MoE includes numerous CNN-based feature extractors, a MoE module for adaptive feature selection, and finally a residual network backbone for deep feature learning. To demonstrate efficacy of our proposed model we evaluated on several datasets, and compared with current state-of-the-art methods. Our model achieves accuracies of 74.77% on AffectNet (v7), 72.55% on AffectNet (v8), 84.29% on RAF-DB, and 64.66% on FER-2013. The results show how adaptive our model is and how it may be used to develop end-to-end emotion recognition systems in practical settings. Reproducible codes and results are made publicly accessible at https://github.com/DeeptimaanB/ExpressNet-MoE.

Large Language Models (LLMs) have significantly advanced the field of natural language processing, enhancing capabilities in both language understanding and generation across diverse domains. However, developing LLMs for Arabic presents unique challenges. This paper explores these challenges by focusing on critical aspects such as data curation, tokenizer design, and evaluation. We detail our approach to the collection and filtration of Arabic pre-training datasets, assess the impact of various tokenizer designs on model performance, and examine the limitations of existing Arabic evaluation frameworks, for which we propose a systematic corrective methodology. To promote transparency and facilitate collaborative development, we share our data and methodologies, contributing to the advancement of language modeling, particularly for the Arabic language.