Agents are predominantly evaluated and optimized via task success metrics, which are coarse, rely on manual design from experts, and fail to reward intermediate emergent behaviors. We propose **AutoLibra**, a framework for agent evaluation, that transforms open-ended human feedback *e.g.* "If you find that the button is disabled, don't click it again", or "This agent has too much autonomy to decide what to do on its own" into metrics for evaluating fine-grained behaviors in agent trajectories. AutoLibra accomplishes this by grounding feedback to an agent's behavior, clustering similar positive and negative behaviors, and creating concrete metrics with clear definitions and concrete examples, which can be used for prompting LLM-as-a-Judge as evaluators. We further propose two meta metrics to evaluate the alignment of a set of (induced) metrics with open feedback: "coverage" and "redundancy". Through optimizing these meta-metrics, we experimentally demonstrate AutoLibra's ability to induce more concrete agent evaluation metrics than the ones proposed in previous agent evaluation benchmarks and discover new metrics to analyze agents. We also present two applications of AutoLibra in agent improvement: First, we show that AutoLibra serve human prompt engineers for diagonalize agent failures and improve prompts iterative. Moreover, we find that AutoLibra can induce metrics for automatic optimization for agents, which makes agents improve through self-regulation. Our results suggest that AutoLibra is a powerful task-agnostic tool for evaluating and improving language agents.

AIs have made rapid progress on research-oriented benchmarks of knowledge and reasoning, but it remains unclear how these gains translate into economic value and automation. To measure this, we introduce the Remote Labor Index (RLI), a broadly multi-sector benchmark comprising real-world, economically valuable projects designed to evaluate end-to-end agent performance in practical settings. AI agents perform near the floor on RLI, with the highest-performing agent achieving an automation rate of 2.5%. These results help ground discussions of AI automation in empirical evidence, setting a common basis for tracking AI impacts and enabling stakeholders to proactively navigate AI-driven labor automation.

As increasingly capable agents are deployed, a central safety question is how to retain meaningful human control without modifying the underlying system. We study a minimal control interface where an agent chooses whether to act autonomously (play) or defer (ask), while a human simultaneously chooses whether to be permissive (trust) or to engage in oversight (oversee). If the agent defers, the human's choice determines the outcome, potentially leading to a corrective action or a system shutdown. We model this interaction as a two-player Markov Game. Our analysis focuses on cases where this game qualifies as a Markov Potential Game (MPG), a class of games where we can provide an alignment guarantee: under a structural assumption on the human's value function, any decision by the agent to act more autonomously that benefits itself cannot harm the human's value. We also analyze extensions to this MPG framework. Theoretically, this perspective provides conditions for a specific form of intrinsic alignment. If the reward structures of the human-agent game meet these conditions, we have a formal guarantee that the agent improving its own outcome will not harm the human's. Practically, this model motivates a transparent control layer with predictable incentives where the agent learns to defer when risky and act when safe, while its pretrained policy and the environment's reward structure remain untouched. Our gridworld simulation shows that through independent learning, the agent and human discover their optimal oversight roles. The agent learns to ask when uncertain and the human learns when to oversee, leading to an emergent collaboration that avoids safety violations introduced post-training. This demonstrates a practical method for making misaligned models safer after deployment.

We introduce the General Incentives-based Framework for Fairness (GIFF), a novel approach for fair multi-agent resource allocation that infers fair decision-making from standard value functions. In resource-constrained settings, agents optimizing for efficiency often create inequitable outcomes. Our approach leverages the action-value (Q-)function to balance efficiency and fairness without requiring additional training. Specifically, our method computes a local fairness gain for each action and introduces a counterfactual advantage correction term to discourage over-allocation to already well-off agents. This approach is formalized within a centralized control setting, where an arbitrator uses the GIFF-modified Q-values to solve an allocation problem. Empirical evaluations across diverse domains, including dynamic ridesharing, homelessness prevention, and a complex job allocation task-demonstrate that our framework consistently outperforms strong baselines and can discover far-sighted, equitable policies. The framework's effectiveness is supported by a theoretical foundation; we prove its fairness surrogate is a principled lower bound on the true fairness improvement and that its trade-off parameter offers monotonic tuning. Our findings establish GIFF as a robust and principled framework for leveraging standard reinforcement learning components to achieve more equitable outcomes in complex multi-agent systems.

While Multi-Agent Systems (MAS) excel at complex tasks, their growing autonomy with operational complexity often leads to critical inefficiencies, such as excessive token consumption and failures arising from misinformation. Existing methods primarily focus on post-hoc failure attribution, lacking proactive, real-time interventions to enhance robustness and efficiency. To this end, we introduce SupervisorAgent, a lightweight and modular framework for runtime, adaptive supervision that operates without altering the base agent's architecture. Triggered by an LLM-free adaptive filter, SupervisorAgent intervenes at critical junctures to proactively correct errors, guide inefficient behaviors, and purify observations. On the challenging GAIA benchmark, SupervisorAgent reduces the token consumption of the Smolagent framework by an average of 29.45% without compromising its success rate. Extensive experiments across five additional benchmarks (math reasoning, code generation, and question answering) and various SoTA foundation models validate the broad applicability and robustness of our approach. The code is available at https://github.com/LINs-lab/SupervisorAgent.

Reinforcement Learning with Verifiable Rewards (RLVR) is a promising approach for enhancing agentic deep search. However, its application is often hindered by low \textbf{Reward Density} in deep search scenarios, where agents expend significant exploratory costs for infrequent and often null final rewards. In this paper, we formalize this challenge as the \textbf{Reward Density Optimization} problem, which aims to improve the reward obtained per unit of exploration cost. This paper introduce \textbf{InfoFlow}, a systematic framework that tackles this problem from three aspects. 1) \textbf{Subproblem decomposition}: breaking down long-range tasks to assign process rewards, thereby providing denser learning signals. 2) \textbf{Failure-guided hints}: injecting corrective guidance into stalled trajectories to increase the probability of successful outcomes. 3) \textbf{Dual-agent refinement}: employing a dual-agent architecture to offload the cognitive burden of deep exploration. A refiner agent synthesizes the search history, which effectively compresses the researcher's perceived trajectory, thereby reducing exploration cost and increasing the overall reward density. We evaluate InfoFlow on multiple agentic search benchmarks, where it significantly outperforms strong baselines, enabling lightweight LLMs to achieve performance comparable to advanced proprietary LLMs.

The theory describes four physical zones (or territories) defined by growing distances around each person, as can be seen in Figure 3 (top left). With those hidden unwritten rules for spaces around a person, only socially close people are welcome within the intimate zone, while generally close people can enter the personal zone, followed by generally familiar people who are allowed in the social space. Otherwise, general public are only permitted within the public space. The concept of group proxemics has been investigated in literature with most attention being paid to detailing the classical proxemics theory. For instance, the authors of [14] explored proxemics and their impact on shape of group formation, the authors of [2] explored proxemics dispersion as average distances people maintain between each other as they walk in group, and in [18], [19] focus was given to studying the effect of proxemics on crowd and its traffic flow dynamics. Within robot-human interactions, the authors of [20]-[22] studied appropriate (safety, comfort, acceptability, etc) distance robots are expected to maintain from people (as individuals). It could be noticed that proxemics are structured around interactions between individuals and details are specified in terms of social relationships between them. In what follows, we explore the situation when an individual is part of a bigger and more complex social entity such as a group. We study the nature of such interactions and and explore associated proxemics.

This article investigates the pedestrian group as an emergent agent. The article explores empirical data to derive emergent agency and formation state spaces and outline recurring patterns of walking behavior. In this analysis, pedestrian trajectories extracted from surveillance videos are used along with manually annotated pedestrian group memberships. We conducted manual expert evaluation of observed groups, produced new manual annotations for relevant events pertaining to group behavior and extracted metrics relevant group formation. This information along with quantitative analysis was used to model the life-cycle and formation of the group agent. Those models give structure to expectations around walking behavior of groups; from pedestrian walking independently to the emergence of a collective intention where group members tended to maintain bounded distance between each other. Disturbances to this bounded distance often happened in association with changes in either their agency or their formation states. We summarized the patterns of behavior along with the sequences of state transitions into abstract patterns, which can aid in the development of more detailed group agents in simulation and in the design of engineering systems to interact with such groups.

Online social networks have transformed the ways in which political mobilization messages are disseminated, raising new questions about how peer influence operates at scale. Building on the landmark 61-million-person Facebook experiment \citep{bond201261}, we develop an agent-based simulation framework that integrates real U.S. Census demographic distributions, authentic Twitter network topology, and heterogeneous large language model (LLM) agents to examine the effect of mobilization messages on voter turnout. Each simulated agent is assigned demographic attributes, a personal political stance, and an LLM variant (\texttt{GPT-4.1}, \texttt{GPT-4.1-Mini}, or \texttt{GPT-4.1-Nano}) reflecting its political sophistication. Agents interact over realistic social network structures, receiving personalized feeds and dynamically updating their engagement behaviors and voting intentions. Experimental conditions replicate the informational and social mobilization treatments of the original Facebook study. Across scenarios, the simulator reproduces qualitative patterns observed in field experiments, including stronger mobilization effects under social message treatments and measurable peer spillovers. Our framework provides a controlled, reproducible environment for testing counterfactual designs and sensitivity analyses in political mobilization research, offering a bridge between high-validity field experiments and flexible computational modeling.\footnote{Code and data available at https://github.com/CausalMP/LLM-SocioPol}

Karl Marx once wrote that ``the human essence is the ensemble of social relations'', suggesting that individuals are not isolated entities but are fundamentally shaped by their interactions with other entities, within which contexts play a constitutive and essential role. With the advent of computers and artificial intelligence, these contexts are no longer limited to purely human--human interactions: human--machine interactions are included as well. Then a central question emerges: How can machines better understand our situations and purposes? To address this challenge, researchers have recently introduced the concept of context engineering. Although it is often regarded as a recent innovation of the agent era, we argue that related practices can be traced back more than twenty years. Since the early 1990s, the field has evolved through distinct historical phases, each shaped by the intelligence level of machines: from early human--computer interaction frameworks built around primitive computers, to today's human--agent interaction paradigms driven by intelligent agents, and potentially to human--level or superhuman intelligence in the future. In this paper, we situate context engineering, provide a systematic definition, outline its historical and conceptual landscape, and examine key design considerations for practice. By addressing these questions, we aim to offer a conceptual foundation for context engineering and sketch its promising future. This paper is a stepping stone for a broader community effort toward systematic context engineering in AI systems.