arXiv subjects

James Evans

Publications and source records attributed to James Evans.

At least 37 records · Page 2Linked to original sources

arXiv2026

Agentic AI and the next intelligence explosion

The "AI singularity" is often miscast as a monolithic, godlike mind. Evolution suggests a different path: intelligence is fundamentally plural, social, and relational. Recent advances in agentic AI reveal that frontier reasoning models, such as DeepSeek-R1, do not improve simply by "thinking longer". Instead, they simulate internal "societies of thought," spontaneous cognitive debates that argue, verify, and reconcile to solve complex tasks. Moreover, we are entering an era of human-AI centaurs: hybrid actors where collective agency transcends individual control. Scaling this intelligence requires shifting from dyadic alignment (RLHF) toward institutional alignment. By designing digital protocols, modeled on organizations and markets, we can build a social infrastructure of checks and balances. The next intelligence explosion will not be a single silicon brain, but a complex, combinatorial society specializing and sprawling like a city. No mind is an island.

cs.AI↗

arXiv2026

Signal in the Noise: Polysemantic Interference Transfers and Predicts Cross-Model Influence

Polysemanticity is pervasive in language models and remains a major challenge for interpretation and model behavioral control. Leveraging sparse autoencoders (SAEs), we map the polysemantic topology of two small models (Pythia-70M and GPT-2-Small) to identify SAE feature pairs that are semantically unrelated yet exhibit interference within models. We intervene at four foci (prompt, token, feature, neuron) and measure induced shifts in the next-token prediction distribution, uncovering polysemantic structures that expose a systematic vulnerability in these models. Critically, interventions distilled from counterintuitive interference patterns shared by two small models transfer reliably to larger instruction-tuned models (Llama-3.1-8B/70B-Instruct and Gemma-2-9B-Instruct), yielding predictable behavioral shifts without access to model internals. These findings challenge the view that polysemanticity is purely stochastic, demonstrating instead that interference structures generalize across scale and family. Such generalization suggests a convergent, higher-order organization of internal representations, which is only weakly aligned with intuition and structured by latent regularities, offering new possibilities for both black-box control and theoretical insight into human and artificial cognition.

cs.AI↗

arXiv2026

Classroom AI: Large Language Models as Grade-Specific Teachers

Large Language Models (LLMs) offer a promising solution to complement traditional teaching and address global teacher shortages that affect hundreds of millions of children, but they fail to provide grade-appropriate responses for students at different educational levels. We introduce a framework for finetuning LLMs to generate age-appropriate educational content across six grade levels, from lower elementary to adult education. Our framework successfully adapts explanations to match students' comprehension capacities without sacrificing factual correctness. This approach integrates seven established readability metrics through a clustering method and builds a comprehensive dataset for grade-specific content generation. Evaluations across multiple datasets with 208 human participants demonstrate substantial improvements in grade-level alignment, achieving a 35.64 percentage point increase compared to prompt-based methods while maintaining response accuracy. AI-assisted learning tailored to different grade levels has the potential to advance educational engagement and equity.

cs.CY↗

arXiv2026

China leads scientific trends; the West launches new ones

How nations shape the scientific frontier matters for technological competition, but standard metrics, including publication counts, citations, and disruption indices, look backward and fail to distinguish between fundamentally different leadership strategies. We develop and validate two forward-looking model-based measures and apply them to tens of millions of articles since 1990. The first embeds research pathways within an evolving hypergraph of concepts and scientists to identify leadership in emerging areas--work that anticipates where the scientific crowd is heading. The second embeds evolving samples of ideas and disciplines drawn upon in past research to identify leadership in surprising new directions as unexpected combinations become routine and science reorganizes around them. China became the global leader in emerging areas roughly a decade ago, well before it led in volume, reflecting a capacity to detect and amplify nascent consensus at scale. The United States and Europe show the opposite profile: declining emergence shares but persistent leadership in prescient work, especially research bridging disciplinary boundaries. These patterns replicate across databases, attribution methods, and strategic domains, including AI, biotechnology, energy, and semiconductors. Nations lead science by reading the landscape or by reshaping it, and the institutional requirements for each strategy lie in tension. The distribution of these strategies promises to shape the global structure of technological leadership for decades.

cs.DL↗

arXiv2026

Urban mobility network centrality predicts social resilience

Cities thrive on social interactions that foster well-being, innovation, and prosperity; yet, exogenous shocks such as pandemics, hurricanes, and wildfires can severely disrupt them. Different urban venues exhibit widely divergent response patterns, raising key questions about what factors contribute to these differences and how we can anticipate and respond. Understanding these questions is crucial for safeguarding social resilience, the capacity of urban venues to maintain both visitation and diversity. In this study, we analyze large-scale human mobility data from 15 US cities covering more than 103 million residents across three distinct urban shocks. Despite a general trend of declining visitation and weakened social mixing, 36.28%-53.01% of venues exhibit reduced segregation, and 21.04%-38.55% of venues exhibit increased visitation. By constructing a mobility network interlinking types of urban venues, we reveal that eigenvector network centrality tends to indicate the provision of essential services and robustly predicts social resilience across varied urban shocks. Specifically, centrality elevates the explanatory power by more than 80% in predicting both segregation and mobility change, compared with more intuitive features. Furthermore, compared to peripheral venues, core venues featuring shorter visit distances, broader neighborhood visitation, shorter visitor dwell times, and steadier popularity throughout the day. Such patterns imply a dual social mechanism: core venues sustain social ties through frequent informal interaction, while peripheral ones facilitate deeper engagement around specialized interests and their corresponding social circles. By bridging urban mobility research with economic theories that distinguish staple from discretionary products, we propose a well-and-pool analogy that suggests how people spend their varying urban mobility budgets.

cs.SI↗

arXiv2026

Research Superalignment Should Advance Now with Alternating Competence and Conformity Optimization

The recent leap in AI capabilities, driven by big generative models, has sparked the possibility of achieving Artificial General Intelligence (AGI) and further triggered discussions on Artificial Superintelligence (ASI)-a system surpassing all humans across measured domains. This gives rise to the critical research question of: As we approach ASI, how do we align it with human values, ensuring it benefits rather than harms human society, a.k.a., the Superalignment problem. Despite ASI being regarded by many as a hypothetical concept, in this position paper, we argue that superalignment is achievable and research on it should advance immediately, through simultaneous and alternating optimization of task competence and value conformity. We posit that superalignment is not merely a safeguard for ASI but also necessary for its responsible realization. To support this position, we first provide a formal definition of superalignment rooted in the gap between capability and capacity, delve into its perceived infeasibility by analyzing the limitations of existing paradigms, and then illustrate a conceptual path of superalignment to support its achievability, centered on two fundamental principles. This work frames a potential initiative for developing value-aligned next-generation AI in the future, which will garner greater benefits and reduce potential harm to humanity.

cs.AI↗

arXiv2026

Generative AI collective behavior needs an interactionist paradigm

In this article, we argue that understanding the collective behavior of agents based on large language models (LLMs) is an essential area of inquiry, with important implications in terms of risks and benefits, impacting us as a society at many levels. We claim that the distinctive nature of LLMs--namely, their initialization with extensive pre-trained knowledge and implicit social priors, together with their capability of adaptation through in-context learning--motivates the need for an interactionist paradigm consisting of alternative theoretical foundations, methodologies, and analytical tools, in order to systematically examine how prior knowledge and embedded values interact with social context to shape emergent phenomena in multi-agent generative AI systems. We propose and discuss four directions that we consider crucial for the development and deployment of LLM-based collectives, focusing on theory, methods, and trans-disciplinary dialogue.

cs.AI↗

arXiv2026

Reasoning Models Generate Societies of Thought

Large language models have achieved remarkable capabilities across domains, yet mechanisms underlying sophisticated reasoning remain elusive. Recent reasoning models outperform comparable instruction-tuned models on complex cognitive tasks, attributed to extended computation through longer chains of thought. Here we show that enhanced reasoning emerges not from extended computation alone, but from simulating multi-agent-like interactions -- a society of thought -- which enables diversification and debate among internal cognitive perspectives characterized by distinct personality traits and domain expertise. Through quantitative analysis and mechanistic interpretability methods applied to reasoning traces, we find that reasoning models like DeepSeek-R1 and QwQ-32B exhibit much greater perspective diversity than instruction-tuned models, activating broader conflict between heterogeneous personality- and expertise-related features during reasoning. This multi-agent structure manifests in conversational behaviors, including question-answering, perspective shifts, and the reconciliation of conflicting views, and in socio-emotional roles that characterize sharp back-and-forth conversations, together accounting for the accuracy advantage in reasoning tasks. Controlled reinforcement learning experiments reveal that base models increase conversational behaviors when rewarded solely for reasoning accuracy, and fine-tuning models with conversational scaffolding accelerates reasoning improvement over base models. These findings indicate that the social organization of thought enables effective exploration of solution spaces. We suggest that reasoning models establish a computational parallel to collective intelligence in human groups, where diversity enables superior problem-solving when systematically structured, which suggests new opportunities for agent organization to harness the wisdom of crowds.

cs.CL↗

arXiv2026

Modularity, Architectural Innovation, and New Venture Success

Startups face a classic dilemma in innovation strategy: should they pursue cumulative, low-risk improvements or disruptive, high-risk breakthroughs? The Henderson and Clark framework suggests that architectural innovation, which reconfigures existing economic modules in novel ways, tends to be disruptive and risky for established organizations, but the success of this strategy for entrepreneurs remains less well understood, largely based on methodological constraints. Building on a complex-economics perspective and advanced computational models, we distinguish architectural innovation from modular innovation, which incrementally updates economic modules, and modular invention, which forges new ones, within the entrepreneurship context. Then we examine how each strategy influences startup performance. We analyze 298,915 U.S. venture-funded start-ups from 1976-2020, embedding company descriptions within a dynamic semantic space constructed from business and patent discourse to measure innovation structure across the entire economy. Event history models reveal that architectural innovation leads to successful IPOs and high-value acquisitions, while both modular innovation and invention increase the risk of failure. By comparing the outcomes of architectural and modular innovation and invention, this paper reveals that what is typically seen as the riskiest form of innovation can, for startups, be the safest route to success. This reconceptualization inverts the trade-off between exploration-exploitation typically assumed in organizational learning with critical implications for entrepreneurial strategy and innovation policy.

econ.EM↗

arXiv2025

Deep versus Broad Technology Search and the Timing of Innovation Impact

This study offers a new perspective on the depth-versus-breadth debate in innovation strategy, by modeling inventive search within dynamic collective knowledge systems, and underscoring the importance of timing for technological impact. Using frontier machine learning to project patent citation networks in hyperbolic space, we analyze 4.9 million U.S. patents to examine how search strategies give rise to distinct temporal patterns in impact accumulation. We find that inventions based on deep search, which relies on a specialized understanding of complex recombination structures, drive higher short-term impact through early adoption within specialized communities, but face diminishing returns as innovations become "locked-in" with limited diffusion potential. Conversely, when inventions are grounded in broad search that spans disparate domains, they encounter initial resistance but achieve wider diffusion and greater long-term impact by reaching cognitively diverse audiences. Individual inventions require both depth and breadth for stable impact. Organizations can strategically balance approaches across multiple inventions: using depth to build reliable technological infrastructure while pursuing breadth to expand applications. We advance innovation theory by demonstrating how deep and broad search strategies distinctly shape the timing and trajectory of technological impact, and how individual inventors and organizations can leverage these mechanisms to balance exploitation and exploration.

physics.soc-ph↗

arXiv2025

Epitome: Pioneering an Experimental Platform for AI-Social Science Integration

Large Language Models (LLMs) enable unprecedented social science experimentation by creating controlled hybrid human-AI environments. We introduce Epitome (www.epitome-ai.com), an open experimental platform that operationalizes this paradigm through Matrix-like social worlds where researchers can study isolated human subjects and groups interacting with LLM agents. This maintains ecological validity while enabling precise manipulation of social dynamics. Epitome approaches three frontiers: (1) methodological innovation using LLM confederates to reduce complexity while scaling interactions; (2) empirical investigation of human behavior in AI-saturated environments; and (3) exploration of emergent properties in hybrid collectives. Drawing on interdisciplinary foundations from management, communication, sociology, psychology, and ethics, the platform's modular architecture spans foundation model deployment through data collection. We validate Epitome through replication of three seminal experiments, demonstrating capacity to generate robust findings while reducing experimental complexity. This tool provides crucial insights for understanding how humans navigate AI-mediated social realities, knowledge essential for policy, education, and human-centered AI design.

cs.CY↗

arXiv2025

Artificial Intelligence Tools Expand Scientists' Impact but Contract Science's Focus (Just accepted by Nature, to be online soon)

Development in Artificial Intelligence (AI) has accelerated scientific discovery. Alongside recent AI-oriented Nobel prizes, these trends establish the role of AI tools in science. This advancement raises questions about the potential influences of AI tools on scientists and science as a whole, and highlights a potential conflict between individual and collective benefits. To evaluate, we used a pretrained language model to identify AI-augmented research, with an F1-score of 0.875 in validation against expert-labeled data. Using a dataset of 41.3 million research papers across natural science and covering distinct eras of AI, here we show an accelerated adoption of AI tools among scientists and consistent professional advantages associated with AI usage, but a collective narrowing of scientific focus. Scientists who engage in AI-augmented research publish 3.02 times more papers, receive 4.84 times more citations, and become research project leaders 1.37 years earlier than those who do not. By contrast, AI adoption shrinks the collective volume of scientific topics studied by 4.63% and decreases scientist's engagement with one another by 22.00%. Thereby, AI adoption in science presents a seeming paradox -- an expansion of individual scientists' impact but a contraction in collective science's reach -- as AI-augmented work moves collectively toward areas richest in data. With reduced follow-on engagement, AI tools appear to automate established fields rather than explore new ones, highlighting a tension between personal advancement and collective scientific progress.

cs.CY↗

arXiv2025

Designing AI-Agents with Personalities: A Psychometric Approach

We introduce a methodology for assigning quantifiable and psychometrically validated personalities to AI-Agents using the Big Five framework. Across three studies, we evaluate its feasibility and limitations. In Study 1, we show that large language models (LLMs) capture semantic similarities among Big Five measures, providing a basis for personality assignment. In Study 2, we create AI-Agents using prompts designed based on the Big Five Inventory-2 (BFI-2) in different format, and find that AI-Agents powered by new models align more closely with human responses on the Mini-Markers test, although the finer pattern of results (e.g., factor loading patterns) were sometimes inconsistent. In Study 3, we validate our AI-Agents on risk-taking and moral dilemma vignettes, finding that models prompted with the BFI-2-Expanded format most closely reproduce human personality-decision associations, while safety-aligned models generally inflate 'moral' ratings. Overall, our results show that AI-Agents align with humans in correlations between input Big Five traits and output responses and may serve as useful tools for preliminary research. Nevertheless, discrepancies in finer response patterns indicate that AI-Agents cannot (yet) fully substitute for human participants in precision or high-stakes projects.

cs.AI↗

arXiv2025

The Strength of Weak Ties Between Open-Source Developers

In a real-world social network, weak ties (reflecting low-intensity, infrequent interactions) act as bridges and connect people to different social circles, giving them access to diverse information and opportunities that are not available within one's immediate, close-knit vicinity. Weak ties can be crucial for creativity and innovation, as they introduce ideas and approaches that people can then combine in novel ways, leading to innovative solutions. Do weak ties facilitate creativity in software in similar ways? This paper suggests that the answer is "yes." Concretely, we study the correlation between developers' knowledge acquisition through three distinct interaction networks on GitHub and the innovativeness of the projects they develop, across over 37,000 Python projects hosted on GitHub. Our findings suggest that the topical diversity of projects in which developers engage, rather than the volume, correlates positively with the innovativeness of their future code. Notably, exposure through weak interactions (e.g., starring) emerges as a stronger predictor of future novelty than via strong ones (e.g., committing)

cs.SE↗

arXiv2025

Hollywood Town: Long-Video Generation via Cross-Modal Multi-Agent Orchestration

Recent advancements in multi-agent systems have demonstrated significant potential for enhancing creative task performance, such as long video generation. This study introduces three innovations to improve multi-agent collaboration. First, we propose OmniAgent, a hierarchical, graph-based multi-agent framework for long video generation that leverages a film-production-inspired architecture to enable modular specialization and scalable inter-agent collaboration. Second, inspired by context engineering, we propose hypergraph nodes that enable temporary group discussions among agents lacking sufficient context, reducing individual memory requirements while ensuring adequate contextual information. Third, we transition from directed acyclic graphs (DAGs) to directed cyclic graphs with limited retries, allowing agents to reflect and refine outputs iteratively, thereby improving earlier stages through feedback from subsequent nodes. These contributions lay the groundwork for developing more robust multi-agent systems in creative tasks.

cs.MA↗

arXiv2025

Topological Structure Learning Should Be A Research Priority for LLM-Based Multi-Agent Systems

Large Language Model-based Multi-Agent Systems (MASs) have emerged as a powerful paradigm for tackling complex tasks through collaborative intelligence. However, the topology of these systems--how agents in MASs should be configured, connected, and coordinated--remains largely unexplored. In this position paper, we call for a paradigm shift toward \emph{topology-aware MASs} that explicitly model and dynamically optimize the structure of inter-agent interactions. We identify three fundamental components--agents, communication links, and overall topology--that collectively determine the system's adaptability, efficiency, robustness, and fairness. To operationalize this vision, we introduce a systematic three-stage framework: 1) agent selection, 2) structure profiling, and 3) topology synthesis. This framework not only provides a principled foundation for designing MASs but also opens new research frontiers across language modeling, reinforcement learning, graph learning, and generative modeling to ultimately unleash their full potential in complex real-world applications. We conclude by outlining key challenges and opportunities in MASs evaluation. We hope our framework and perspectives offer critical new insights in the era of agentic AI.

cs.MA↗

arXiv2025

Rationality Check! Benchmarking the Rationality of Large Language Models

Large language models (LLMs), a recent advance in deep learning and machine intelligence, have manifested astonishing capacities, now considered among the most promising for artificial general intelligence. With human-like capabilities, LLMs have been used to simulate humans and serve as AI assistants across many applications. As a result, great concern has arisen about whether and under what circumstances LLMs think and behave like real human agents. Rationality is among the most important concepts in assessing human behavior, both in thinking (i.e., theoretical rationality) and in taking action (i.e., practical rationality). In this work, we propose the first benchmark for evaluating the omnibus rationality of LLMs, covering a wide range of domains and LLMs. The benchmark includes an easy-to-use toolkit, extensive experimental results, and analysis that illuminates where LLMs converge and diverge from idealized human rationality. We believe the benchmark can serve as a foundational tool for both developers and users of LLMs.

cs.AI↗

arXiv2025

We Need a New Ethics for a World of AI Agents

The deployment of capable AI agents raises fresh questions about safety, human-machine relationships and social coordination. We argue for greater engagement by scientists, scholars, engineers and policymakers with the implications of a world increasingly populated by AI agents. We explore key challenges that must be addressed to ensure that interactions between humans and agents, and among agents themselves, remain broadly beneficial.

cs.CY↗