arXiv subjects

Dashun Wang

Publications and source records attributed to Dashun Wang.

At least 19 recordsLinked to original sources

arXiv2026

The Legibility Gap: How Gender Equity Interventions Redistribute Recognition Across Cultures

Efforts to promote gender equity in science increasingly rely on name-based inference to quantify representation and guide policy and behavior. Yet linguistic cues that signal gender vary across cultures and are often obscured when names are transliterated into English. Here we identify a pattern we call the "legibility gap": when gender is inferred from names, equity interventions systematically benefit women whose names signal gender while bypassing those whose names lose such cues in translation. Using both observational and experimental evidence, we show how this gap reshapes recognition in science. Analyzing citation diversity statements-an emerging practice in which authors report the algorithmically estimated gender composition of their reference lists-we find that papers that include this practice cite women more frequently, but the gains accrue almost entirely to authors with gender-signaling Western names. By contrast, women whose names lose gender cues in English transliteration, predominantly those with East Asian names, receive fewer citations in these same papers. Two preregistered experiments (N = 2,250) corroborate this pattern and identify its mechanism: linguistic legibility, not cultural unfamiliarity, determines who is recognized as a woman and who benefits from policies designed to support women in science. Overall, these findings expose a previously unrecognized layer of inequity embedded in global equity infrastructures. As science becomes increasingly global and equity efforts increasingly algorithmic, the legibility gap reveals how uneven identity recognition reshapes fairness. In global systems of recognition, equity depends not only on whether policies are effective on average, but also on whether they are equitable across cultures.

cs.DL↗

arXiv2026

The Production of Policy Knowledge in the United States

Policymaking relies on institutions that translate expertise into politically actionable knowledge. Yet little is known about how such knowledge is created, structured, and politically polarized, or how science shapes these processes. Using 2 million U.S. policy documents and nearly 1 million scientific citations from more than 200 think tanks and 100 government organizations, we analyze the supply side of science in policymaking: the production of policy knowledge. We find that think tanks are the dominant suppliers of science-based policy knowledge to government, but that this production has become increasingly politically polarized, driven primarily by growing insularity among left-leaning institutions. We further find that science is linked to lower polarization, as policy documents grounded in scientific evidence, especially those citing high-impact science, are less ideologically segregated and occupy more central positions in policy knowledge networks. These results reveal the potential of scientific expertise to shape and, at times, bridge the ideological landscape of policy knowledge production. Amid rising political polarization and the growing role of science in policymaking, understanding how policy knowledge is produced and how scientific expertise relates to its ideological structure is essential to strengthening the informational foundations of democratic governance.

physics.soc-ph↗

arXiv2026

Human-AI Collaboration in Science at Scale: A Global Large-scale Randomized Field Experiment

Collaboration is the defining mode of modern science, yet its core mechanism -- feedback -- remains hard to observe, difficult to scale, and unequally distributed. Here we test whether large language models (LLMs) can contribute to this hidden but vital practice and reallocate scientific feedback, an essential yet scarce resource for knowledge production. In a global large-scale randomized field experiment, we delivered customized LLM-generated feedback for over 31,000 arXiv preprints across 150 fields and more than 45,000 researchers from 133 geographic regions. Relative to controls, authors who received feedback had a significantly higher likelihood of revising their manuscripts, corresponding to a 12.55% relative increase over the baseline revision rate. Exposure to AI feedback also increased authors' subsequent use of LLM tools in their future papers, suggesting longer-run shifts in scientific practice. These effects were strongest among authors from non-English-dominant research regions, manuscripts less embedded in the scholarly literature, and teams with lower h-indexes and earlier career stages, consistent with the idea that AI feedback may provide the greatest benefit where access to timely critique is otherwise limited. Together, these findings provide causal evidence that structured AI-based interventions can transform access to scientific feedback from a largely private advantage into a more widely distributed resource, with broader implications for productivity, equity, and capacity across the global research system.

physics.soc-ph↗

arXiv2026

Universal Dynamics of Punctuated Progress

Scientific and technological frontiers advance through punctuated dynamics, yet the principles governing these dynamics remain poorly understood. Here we collect and analyze datasets tracking the evolution of frontiers across 9 different domains, spanning materials discovery, structural biology, AI, computational biomedicine, data science, theoretical computer science, Formula-1 racing, and physical wheel building. Analyzing 6.8M solutions to 6.7K tasks, we uncover three universal patterns: (1) waiting times between new frontiers are heavy-tailed, with most attempts concentrated in long stasis; (2) frontier records accumulate at a sublinear rate, faster than logarithmic yet slower than linear growth; (3) record-breaking events are temporally correlated, generating short-term predictability yet long-term unpredictability. Despite the differences in the scale, scope, and definition of the settings, these patterns are remarkably consistent across all domains we study, and are not captured by models from complex systems, record statistics, economics of innovation, and cultural evolution. We trace the missing ingredient to the distinction between radical and incremental innovation, and develop a minimal, analytically solvable model incorporating both radical resets that restructure what is achievable and incremental refinements that exploit the current frontier. The simple model reproduces all three empirical regularities. Remarkably, the leading-order predictions are parameter-independent, identifying a new universality class governing punctuated progress and yielding testable predictions about how openness and access to frontier solutions shape the pace of advance. Overall, these results reveal universal dynamics governing punctuated progress and identify the interplay between radical resets and incremental refinements as the key driver of how scientific and technological frontiers advance.

physics.soc-ph↗

arXiv2026

Figures as Interfaces: Toward LLM-Native Artifacts for Scientific Discovery

Large language models (LLMs) are transforming scientific workflows, not only through their generative capabilities but also through their emerging ability to use tools, reason about data, and coordinate complex analytical tasks. Yet in most human-AI collaborations, the primary outputs, figures, are still treated as static visual summaries: once rendered, they are handled by both humans and multimodal LLMs as images to be re-interpreted from pixels or captions. The emergent capabilities of LLMs open an opportunity to fundamentally rethink this paradigm. In this paper, we introduce the concept of LLM-native figures: data-driven artifacts that are simultaneously human-legible and machine-addressable. Unlike traditional plots, each artifact embeds complete provenance: the data subset, analytical operations and code, and visualization specification used to generate it. As a result, an LLM can "see through" the figure--tracing selections back to their sources, generating code to extend analyses, and orchestrating new visualizations through natural-language instructions or direct manipulation. We implement this concept through a hybrid language-visual interface that integrates LLM agents with a bidirectional mapping between figures and underlying data. Using the science of science domain as a testbed, we demonstrate that LLM-native figures can accelerate discovery, improve reproducibility, and make reasoning transparent across agents and users. More broadly, this work establishes a general framework for embedding provenance, interactivity, and explainability into the artifacts of modern research, redefining the figure not as an end product, but as an interface for discovery. For more details, please refer to the demo video available at www.llm-native-figure.com.

cs.HC↗

arXiv2026

Benchmark Illusion: Disagreement among LLMs and Its Scientific Consequences

Benchmarks underpin how progress in large language models (LLMs) is measured and trusted. Yet our analyses reveal that apparent convergence in benchmark accuracy can conceal deep epistemic divergence. Using two major reasoning benchmarks - MMLU-Pro and GPQA - we show that LLMs achieving comparable accuracy still disagree on 16-66% of items, and 16-38% among top-performing frontier models. These discrepancies suggest distinct error profiles for different LLMs. When such models are used for scientific data annotation and inference, their hidden disagreements propagate into research results: in re-analyses of published studies in education and political science, switching the annotation model can change estimated treatment effects by more than 80%, and in some cases reverses their sign. Together, these findings illustrate a benchmark illusion, where equal accuracy may conceal disagreement, with model choice becoming a hidden yet consequential variable for scientific reproducibility.

cs.CL↗

arXiv2026

Pivoting as an Adaptive Strategy to Geopolitical Tensions in U.S. Science

Geopolitical tensions increasingly reshape the structure and openness of global science, yet we still lack a clear understanding of how successfully scientists adapt their work under such pressures. Using millions of funding and publication datasets across the past ten years, we investigate how U.S. China geopolitical tensions reshaped individual research activities of U.S. based scientists, particularly those collaborating with Chinese peers. We find that although U.S. China geopolitical tensions significantly reduce funding opportunities, many scientists actively respond by pivoting their research portfolios toward alternative topics, and this adaptive reorientation partially mitigates funding losses. Crucially, the effectiveness of this adaptive strategy is highly unequal: for scientists in high risk domains, those of Asian descent, and early-career scientists, pivoting offers only limited protection against funding loss. Our results demonstrate that geopolitical tensions reshape science through shifts in scientists' strategic decisions about their research focus. Understanding this adaptive but uneven reconfiguration is essential for science policies to strengthen the resilience and inclusiveness of the scientific enterprise.

physics.soc-ph↗

arXiv2026

The Rise of Large Language Models and the Direction and Impact of US Federal Research Funding

Federal research funding shapes the direction, diversity, and impact of the US scientific enterprise. Large language models (LLMs) are rapidly diffusing into scientific practice, holding substantial promise while raising widespread concerns. Despite growing attention to AI use in scientific writing and evaluation, little is known about how the rise of LLMs is reshaping the public funding landscape. Here, we examine LLM involvement at key stages of the federal funding pipeline by combining two complementary data sources: confidential National Science Foundation (NSF) and National Institutes of Health (NIH) proposal submissions from two large US R1 universities, including funded, unfunded, and pending proposals, and the full population of publicly released NSF and NIH awards. We find that LLM use rises sharply beginning in 2023 and exhibits a bimodal distribution, indicating a clear split between minimal and substantive use. Across both private submissions and public awards, higher LLM involvement is consistently associated with lower semantic distinctiveness, positioning projects closer to recently funded work within the same agency. The consequences of this shift are agency-dependent. LLM use is positively associated with proposal success and higher early-stage publication output at NIH, whereas no comparable associations are observed at NSF. Notably, the productivity gains at NIH are concentrated in non-hit papers rather than the most highly cited work. Together, these findings provide large-scale evidence that the rise of LLMs is reshaping how scientific ideas are positioned, selected, and translated into publicly funded research, with implications for portfolio governance, research diversity, and the long-run impact of science.

cs.DL↗

arXiv2025

Sci2Pol: Evaluating and Fine-tuning LLMs on Scientific-to-Policy Brief Generation

We propose Sci2Pol-Bench and Sci2Pol-Corpus, the first benchmark and training dataset for evaluating and fine-tuning large language models (LLMs) on policy brief generation from a scientific paper. We build Sci2Pol-Bench on a five-stage taxonomy to mirror the human writing process: (i) Autocompletion, (ii) Understanding, (iii) Summarization, (iv) Generation, and (v) Verification. It features 18 tasks in multiple-choice and open-ended formats. Specifically, for the Generation stage, we show that BERTScore and ROUGE scores fail to capture the quality of brief writing, and introduce a new LLM-based evaluation metric aligned with expert judgement. Using this benchmark, we evaluate 13 leading open-source and commercial LLMs to uncover key limitations. To improve LLM performance on brief writing, we curate the Sci2Pol-Corpus for fine-tuning. We start by linking each cited scientific paper to its corresponding policy document, drawn from 5.6 million policy records. This produces 140,000 candidate pairs. We then employ an LLM-as-a-judge to filter high-quality examples, followed by in-context polishing using three expert-written samples as references. This process yields a final set of 639 new pairs. Finally, we fine-tune three models on Sci2Pol-Corpus: LLaMA-3.18B, Gemma-12B, and Gemma-27B. Fine-tuning leads to consistent performance improvements across Sci2Pol-Bench. Notably, after fine-tuning, Gemma-27B surpasses the much larger GPT-4o and DeepSeek-V3 (671B). These demonstrate the effectiveness of our corpus in bridging the gap between science and policy.

cs.CE↗

arXiv2025

Funding the Frontier: Visualizing the Broad Impact of Science and Science Funding

Understanding the broad impact of science and science funding is critical to ensuring that science investments and policies align with societal needs. Existing research links science funding to the output of scientific publications but largely leaves out the downstream uses of science and the myriad ways in which investing in science may impact human society. As funders seek to allocate scarce funding resources across a complex research landscape, there is an urgent need for informative and transparent tools that allow for comprehensive assessments and visualization of the impact of funding. Here we present Funding the Frontier (FtF), a visual analysis system for researchers, funders, policymakers, university leaders, and the broad public to analyze multidimensional impacts of funding and make informed decisions regarding research investments and opportunities. The system is built on a massive data collection that connects 7M research grants to 140M scientific publications, 160M patents, 10.9M policy documents, 800K clinical trials, and 5.8M newsfeeds, with 1.8B citation linkages among these entities, systematically linking science funding to its downstream impacts. As such, Funding the Frontier is distinguished by its multifaceted impact analysis framework. The system incorporates diverse impact metrics and predictive models that forecast future investment opportunities into an array of coordinated views, allowing for easy exploration of funding and its outcomes. We evaluate the effectiveness and usability of the system using case studies and expert interviews. Feedback suggests that our system not only fulfills the primary analysis needs of its target users, but the rich datasets of the complex science ecosystem and the proposed analysis framework also open new avenues for both visualization and the science of science research.

cs.HC↗

arXiv2025

Peer Review and the Diffusion of Ideas

This study examines a fundamental yet overlooked function of peer review: its role in exposing reviewers to new and unexpected ideas. Leveraging a natural experiment involving over half a million peer review invitations covering both accepted and rejected manuscripts, and integrating high-scale bibliographic and editorial records for 37,279 submitting authors, we find that exposure to a manuscript's core ideas significantly influences the future referencing behavior and knowledge of reviewer invitees who decline the review invite. Specifically, declining reviewer invitees who could view concise summaries of the manuscript's core ideas not only increase their citations to the manuscript itself but also demonstrate expanded breadth, depth, diversity, and prominence of citations to the submitting author's broader body of work. Overall, these results suggest peer review substantially influences the spread of scientific knowledge. Ironically, while the massive scale of peer review, entailing millions of reviews annually, often drives policy debates about its costs and burdens, our findings demonstrate that precisely because of this scale, peer review serves as a powerful yet previously unrecognized engine for idea diffusion, which is central to scientific advances and scholarly communication.

physics.soc-ph↗

arXiv2025

Synthesis of innovation and obsolescence

Innovation and obsolescence describe the dynamics of ever-churning social and biological systems, from the development of economic markets to scientific and technological progress to biological evolution. They have been widely discussed, but in isolation, leading to fragmented modeling of their dynamics. This poses a problem for connecting and building on what we know about their shared mechanisms. Here we collectively propose a conceptual and mathematical framework to transcend field boundaries and to explore unifying theoretical frameworks and open challenges. We ring an optimistic note for weaving together disparate threads with key ideas from the wide and largely disconnected literature by focusing on the duality of innovation and obsolescence and by proposing a mathematical framework to unify the metaphors between constitutive elements.

physics.soc-ph↗

arXiv2025

SciSciGPT: Advancing Human-AI Collaboration in the Science of Science

The increasing availability of large-scale datasets has fueled rapid progress across many scientific fields, creating unprecedented opportunities for research and discovery while posing significant analytical challenges. Recent advances in large language models (LLMs) and AI agents have opened new possibilities for human-AI collaboration, offering powerful tools to navigate this complex research landscape. In this paper, we introduce SciSciGPT, an open-source, prototype AI collaborator that uses the science of science as a testbed to explore the potential of LLM-powered research tools. SciSciGPT automates complex workflows, supports diverse analytical approaches, accelerates research prototyping and iteration, and facilitates reproducibility. Through case studies, we demonstrate its ability to streamline a wide range of empirical and analytical research tasks while highlighting its broader potential to advance research. We further propose an LLM Agent capability maturity model for human-AI collaboration, envisioning a roadmap to further improve and expand upon frameworks like SciSciGPT. As AI capabilities continue to evolve, frameworks like SciSciGPT may play increasingly pivotal roles in scientific research and discovery, unlocking further opportunities. At the same time, these new advances also raise critical challenges, from ensuring transparency and ethical use to balancing human and AI contributions. Addressing these issues may shape the future of scientific inquiry and inform how we train the next generation of scientists to thrive in an increasingly AI-integrated research ecosystem.

cs.AI↗

arXiv2024

Collective dynamics behind success

Understanding the collective dynamics behind the success of ideas, products, behaviors, and social actors is critical for decision-making across diverse contexts, including hiring, funding, career choices, and the design of interventions for social change. Methodological advances and the increasing availability of big data now allow for a broader and deeper understanding of the key facets of success. Recent studies unveil regularities beneath the collective dynamics of success, pinpoint underlying mechanisms, and even enable predictions of success across diverse domains, including science, technology, business, and the arts. However, this research also uncovers troubling biases that challenge meritocratic views of success. This review synthesizes the growing, cross-disciplinary literature on the collective dynamics behind success and calls for further research on cultural influences, the origins of inequalities, the role of algorithms in perpetuating them, and experimental methods to further probe causal mechanisms behind success. Ultimately, these efforts may help to better align success with desired societal values.

physics.soc-ph↗

arXiv2024

Correlated Growth of Causal Networks

The study of causal structure in complex systems has gained increasing attention, with many recent studies exploring causal networks that capture cause-effect relationships across diverse fields. Despite increasing empirical evidence linking causal structures to network topological correlations, the mechanisms underlying the emergence of these correlations in causal networks remain poorly understood. In this work, we propose a general growth framework for causal networks, incorporating two key types of correlations: causal and dynamic. We analytically demonstrate that degree correlations emerge as a consequence of marginal dependencies on these correlations. Our theoretical predictions align quantitatively with empirical data from four large-scale innovation networks. Our theory not only sheds light on the origins of topological correlations but also provides a general framework for understanding correlated growth across causal systems.

physics.soc-ph↗

arXiv2024

Quantifying the Dynamics of Innovation Abandonment Across Scientific, Technological, Commercial, and Pharmacological Domains

Despite the vast literature on the diffusion of innovations that impacts a broad range of disciplines, our understanding of the abandonment of innovations remains limited yet is essential for a deeper understanding of the innovation lifecycle. Here, we analyze four large-scale datasets that capture the temporal and structural patterns of innovation abandonment across scientific, technological, commercial, and pharmacological domains. The paper makes three primary contributions. First, across these diverse domains, we uncover one simple pattern of preferential abandonment, whereby the probability for individuals or organizations to abandon an innovation increases with time and correlates with the number of network neighbors who have abandoned the innovation. Second, we find that the presence of preferential abandonment fundamentally alters the way in which the underlying ecosystem breaks down, inducing a novel structural collapse in networked systems commonly perceived as robust against abandonments. Third, we derive an analytical framework to systematically understand the impact of preferential abandonment on network dynamics, pinpointing specific conditions where it may accelerate, decelerate, or have an identical effect compared to random abandonment, depending on the network topology. Together, these results deepen our quantitative understanding of the abandonment of innovation within networked social systems, with implications for the robustness and functioning of innovation communities. Overall, they demonstrate that the dynamics of innovation abandonment follow simple yet reproducible patterns, suggesting that the uncovered preferential abandonment may be a generic property of the innovation lifecycle.

physics.soc-ph↗

arXiv2024

Tenure and Research Trajectories

Tenure is a cornerstone of the US academic system, yet its relationship to faculty research trajectories remains poorly understood. Conceptually, tenure systems may act as a selection mechanism, screening in high-output researchers; a dynamic incentive mechanism, encouraging high output prior to tenure but low output after tenure; and a creative search mechanism, encouraging tenured individuals to undertake high-risk work. Here, we integrate data from seven different sources to trace US tenure-line faculty and their research outputs at an unprecedented scale and scope, covering over 12,000 researchers across 15 disciplines. Our analysis reveals that faculty publication rates typically increase sharply during the tenure track and peak just before obtaining tenure. Post-tenure trends, however, vary across disciplines: in lab-based fields, such as biology and chemistry, research output typically remains high post-tenure, whereas in non-lab-based fields, such as mathematics and sociology, research output typically declines substantially post-tenure. Turning to creative search, faculty increasingly produce novel, high-risk research after securing tenure. However, this shift toward novelty and risk-taking comes with a decline in impact, with post-tenure research yielding fewer highly cited papers. Comparing outcomes across common career ages but different tenure years or comparing research trajectories in tenure-based and non-tenure-based research settings underscores that breaks in the research trajectories are sharply tied to the individual's tenure year. Overall, these findings provide a new empirical basis for understanding the tenure system, individual research trajectories, and the shape of scientific output.

physics.soc-ph↗

arXiv2024

Adaptability and the Pivot Penalty in Science and Technology

Scientists and inventors set the direction of their work amidst an evolving landscape of questions, opportunities, and challenges. This paper introduces a measurement framework to quantify how far researchers move from their existing research when producing new works. We apply this framework to millions of scientific publications and patents and uncover a pervasive "pivot penalty", where the impact of new research steeply declines the further a researcher moves from their prior work. The pivot penalty applies nearly universally across scientific publishing and patenting and has been growing in magnitude over the past five decades. While creativity frameworks suggest a benefit to exploratory search by researchers and often emphasize outsider advantages in driving breakthroughs, we find little evidence for such an advantage. The pivot penalty is consistent with increasingly narrow specializations of researchers, and when researchers undertake large pivots, a signature of their work is weak engagement with established mixtures of prior knowledge. Unexpected shocks to the research landscape, which may push researchers away from existing areas or pull them into new ones, further demonstrate substantial pivot penalties. COVID-19 provides a high-scale case study, where many researchers engaged the pandemic, yet the pivot penalty remains severe. The pivot penalty generalizes across fields, career stage, productivity, collaboration, and funding contexts, highlighting both the breadth and depth of the adaptive challenge. Overall, the findings point to large and increasing challenges in adapting to new opportunities and threats. The results have implications for individual researchers, research organizations, science policy, and the capacity of science and society as a whole to confront emergent demands.

cs.DL↗