Named after the hundred-eyed watchman of Greek myth, Argus watches the education landscape: spotting new opportunities, pressure-testing the ventures we're building, and tracing every read back to the real-world signals behind it.
The evidence library: the raw signals the pipeline is watching across the education ecosystem. Every idea is built from these.
arXiv:2607.06074v1 Announce Type: cross Abstract: Prompt engineering has emerged as a critical yet undertaught skill for software developers, one that traditional learning approaches are ill-equipped to support given its evolving, interactive, and context-dependent nature. In this paper, we introduce Prompt Coach (PC), an agentic tutor that helps developers learn how to craft high-quality code-generation prompts through Socratic guidance embedded in-flow within their IDE. PC evaluates prompt quality across multiple dimensions and surfaces targeted questions to guide self-correction, grounded in the developer's codebase and the behavior of the target LLM. We present an early empirical study with 15 professional developers combining quantitative prompt quality scoring with qualitative perception measures. Participants showed statistically significant improvements after a single 60-minute session, with the largest gains across dimensions commonly overlooked by developers. They also report
arXiv:2607.05871v1 Announce Type: cross Abstract: Debugging exercises are often assessed from final code and test outcomes, yet these artifacts hide how students reproduced failures, formed hypotheses, inspected evidence, edited code, and verified fixes. We present DebugTracker, a Visual Studio Code extension that records lightweight debugging-process evidence for classroom tasks. DebugTracker separates uncoached Evaluation Mode traces from coached Training Mode traces, stores append-only JSONL events, and exports timeline and Markdown reports for human review. The prototype records test commands, editor and debugger metadata, student checkpoints, source snapshots, optional image evidence, human labels, and optional AI-assisted practice feedback. DebugTracker is largely language-agnostic: it captures process evidence through standard VS Code mechanisms rather than language-specific tooling, although debugger evidence depends on the relevant VS Code language extension. We validate the p
arXiv:2607.05552v1 Announce Type: cross Abstract: Large language models (LLMs) increasingly issue judgments read as binary verdicts, and a growing literature reports such judgments shifting under logically irrelevant changes of wording - among them an amplified yes-no bias on moral dilemmas, absent in humans. A single framing cannot say what such a shift is: in a yes/no question the word "no" is at once logical verdict, lexical token, and last-printed option. We introduce a psychometric battery that separates these: crossed symmetrization - every logically irrelevant factor flipped in balanced pairs - across a corpus of question forms. A graded rating across logically equivalent forms recovers a coherent internal moral scale: frontier models' stance $\theta$ is nearly format-invariant (cross-form incoherence 0.12-0.21 on a $\pm 1$ axis); small open-weight models fail in model-specific ways. Forcing the verdict through yes/no overlays a decomposable artifact: an order bias toward the la
arXiv:2607.05454v1 Announce Type: cross Abstract: The journal is a seventeenth-century technology asked to do four modern jobs at once: disseminate results, certify their quality, allocate scholarly attention, and confer career credit. It does none of them well. Pre-publication peer review is slow, only weakly reliable, demonstrably biased toward established authors and institutions, and expensive, while the reviewing effort it consumes is spent largely on work that will never matter. We argue that these are not defects to be patched but consequences of bundling dissemination and certification into a single gated act, and we propose unbundling them. Under the proposal, authors deposit papers in an open archive; certification happens \emph{after} deposit, continuously, through attributed and up- or down-voted public commentary to which authors may reply; and papers are version-controlled objects that any qualified reader may \emph{fork}, so that the lineage of an idea -- and hence the c
arXiv:2607.05434v1 Announce Type: cross Abstract: Artificial Intelligence (AI) models, at their core, apply general learnings from broad datasets to individual circumstances using probabilistic behaviour. This inductive approach stands in contrast to deductive reasoning approaches which seek to prove conclusions from their premises. However, research has shown that deductive reasoning with AI models is a challenging problem and in the real-world it may not always be feasible. An alternative way forward is to leverage abductive reasoning, seeking to corroborate the output of multiple approaches to identify the most likely conclusion from the factual matrix. We apply this to synthetic media detection in forensic settings, and find we are able to disproportionately lower the risk of false positives to true positive recall. We also provide the first empirical evaluation of OpenAI's rollout of SynthID on synthetic images and evaluate how complementary different synthetic media detection app
arXiv:2607.05420v1 Announce Type: cross Abstract: This study examines how alternative systems of scholarly representation identify and characterize broad public administration (PA) and artificial intelligence related public administration (AI-in-PA) scholarship. Using Web of Science and OpenAlex, it compares five approaches based on author-defined, citation-driven, and AI-assisted representations. The results highlight substantial differences in corpus size, publication types, publishing outlets, temporal development, and thematic clustering and structure. The alternative approaches often identify different knowledge domains instead of varied subsets of the same scholarship and therefore produce distinct representations, as evidenced by no overlap in publications and publishing outlets across representations. The findings suggest that algorithmic knowledge organization increasingly influences how interdisciplinary scholarship is classified, structured, and understood and, epistemologic
arXiv:2607.05398v1 Announce Type: cross Abstract: Personas are often employed to guide large language model agents, yet their effectiveness in shaping strategic behavior in social dilemma settings remains uncertain. To address this, we examined the impact of persona prompts in an iterated Split or Steal game where persona-driven agents interacted with a Virtual Human (VH) controlled by a fixed prompt. Agents were instantiated from four open models (Ministral 3:3b, phi4:14b, Gemma3:12b, and Gemma4:e4b) at two temperature settings (0.3 and 0.7) and deterministic decision with zero temperature, while the VH was powered by GPT 4.1 mini. Across 160 sessions of 15 rounds each conducted in European Portuguese, mutual Split outcomes dominated (roughly 74 percent of rounds), with exploitation occurring in fewer than 11 percent of rounds. Model choice significantly influenced behavior: phi4 and Ministral 3:3b remained consistently cooperative across temperatures, whereas Gemma3:12b and Gemma4:e4
arXiv:2607.06260v1 Announce Type: new Abstract: Large language models (LLMs) are emerging as a new informational layer over cities, shaping which places people discover, consider, and ultimately visit. Yet little is known about which places they surface, which they ignore, and whether these patterns vary across communities and users and translate into real-world economic consequences. Here, we audit restaurant recommendations from three major LLMs across 304 neighborhoods in five U.S. cities using 320 synthetic user profiles spanning income, age, sex, and residential status. We find that LLMs both fabricate venues and systematically overlook real ones. Fabrication is concentrated in neighborhoods with weaker digital and physical footprints and disappears when models are provided with verified venue lists. In contrast, invisibility persists: even when choosing from a fixed set of real venues, 47.5% of establishments are never recommended, and 31.9% of these blind spots are shared across
arXiv:2607.05808v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly integrated into computing education, yet nearly all prior research has focused on text-based interactions. As voice-enabled interfaces become more capable and more common, there is growing interest in understanding how voice input might shape students' use of LLM-powered tools. In this exploratory study, we investigated how introductory programming students interact with Prompt Problems, which are programming tasks that require crafting natural-language prompts to generate correct code. Students (N = 919) solved a series of Prompt Problems with the freedom to select or switch between text and voice input modalities. We collected their prompt submissions as well as post-activity survey responses, then analysed differences in prompt accuracy, persistence, and perspectives by modality. For two of the three problems, we found that students who typed their prompts using text were more likely to hav
arXiv:2607.05680v1 Announce Type: new Abstract: AI systems are increasingly used to provide legal advice, raising questions about whether laypeople accept guidance from algorithms--especially when that advice is legally correct but socially controversial. We report a preregistered survey experiment with 3,348 adults in mainland China examining how people evaluate identical legal advice when it is attributed either to an AI system or to a human lawyer, and when it is accompanied by reasoning or not. Contrary to expectations of algorithm aversion, attribution to an AI system has no net effect on perceived reasonableness. However, mediation analyses reveal opposing psychological pathways underlying this null result. AI-attributed advice is perceived as more objective, which increases perceived reasonableness, but also as less comprehensive and less attentive to special circumstances, which decreases perceived reasonableness. By contrast, providing legal reasoning substantially increases p
arXiv:2607.05574v1 Announce Type: new Abstract: Artificial intelligence increasingly mediates consequential decisions in healthcare, law, and public services, and the field has responded with an extensive methodology for measuring and mitigating bias. Yet the fairness definitions, benchmarks, and debiasing frameworks on which this methodology rests are treated as universal while being produced by a research community whose composition has never been characterized. We show that the AI bias research are structurally concentrated, and that this concentration is greatest, geographically, in precisely the domain the rest of the field inherits from. Analyzing 692 publications spanning five thematic domains, combining bibliometric analysis with semantic clustering, we find that research activity is dominated by a small set of countries, institutions, and authors, with the United States leading publication output and collaboration networks across every domain and most strongly in general fairn
arXiv:2607.05413v1 Announce Type: new Abstract: Open Source Software (OSS) forms a critical layer of contemporary digital infrastructure, yet remains largely overlooked by the institutions and societies that depend on it. Despite growing institutional interest, the causal impact of public funding on OSS project sustainability remains empirically unresolved. Existing literature is divided between econometric and socio-technical approaches with few attempts at causal identification. This work aims to bridge that divide by combining a Goal-Question-Metric framework with the Generalized Synthetic Control Method to estimate the causal effect of the Sovereign Tech Fund on OSS repository activity. Counterfactual trajectories are constructed from a matched donor pool of unfunded projects, enabling identification of what funded repositories would have looked like in the absence of intervention. The main results show that the funding has a significant positive effect on project velocity metrics:
arXiv:2607.05412v1 Announce Type: new Abstract: STEM education faces challenges in personalization and interdisciplinary integration. AI technology has brought new possibilities, but the mechanisms by which AI reshapes the STEM education ecosystem require systematic investigation. This study employs bibliometric methods to analyze 242 publications from 2015-2025, constructing knowledge maps to reveal the evolutionary trajectory. The findings show that the field has transformed from intelligent tutoring systems to inquiry-based learning and computational thinking cultivation driven by LLMs. AI's key contribution lies in providing intelligent scaffolding that lowers the threshold for understanding knowledge. In this sense, AI is a core driving force promoting its shift from knowledge transmission to capability development.
arXiv:2607.05411v1 Announce Type: new Abstract: Higher education institutions are increasingly expected to ensure that both students and staff develop Generative AI (GenAI) literacies. In response, they are introducing professional development programs and embedding GenAI skills within student curricula. However, current educational frameworks typically assume a linear progression of GenAI literacy, implying that foundational technical understanding must precede creative application. This paper challenges such an assumption through a psychometric analysis of a taxonomy-based self-assessment instrument (n = 158). We applied Rasch measurement theory and Guttman ordering to map the latent perceived order of difficulty of GenAI skills across students, academics, and professional staff. Results reveal a fundamental divergence in perceived competence profiles: while academics follow a more traditional linear path, students exhibit an "inverted" profile, frequently mastering high-level creati
arXiv:2607.05410v1 Announce Type: new Abstract: We present CANONIC: governed intelligence that compiles digital artifacts into an evidence ledger at scale. Large language models generate prose faster than anyone can check it, the failure Oxford Languages named 'slop', its 2025 Word of the Year. CANONIC governs whether content may enter a corpus the way a compiler decides whether a program is well-formed: mechanically, by a grammar, at the boundary of admission. Governance reduces to three axioms (Triad, Inheritance, Introspection) that map one-to-one onto compiler theory's syntax, scope-resolution, and type-system layers, and admission is a decidable, linear-time check. We then ask, with a pre-registered cross-provider benchmark across four regimes, whether structural admission keeps slop out. It does not: no prose-reading gate reliably separates reliable from unreliable content. Slop is not a property an algorithm computes. It is a verdict of domain expertise. So a governance layer do
arXiv:2607.05409v1 Announce Type: new Abstract: Introductory programming instruction relies on hands-on practice and short learning activities to support mastery of foundational concepts. Although many such learning resources exist, organizing and linking these items in instructionally meaningful ways is challenging without time-intensive expert curation. This study investigates the use of pattern-based Knowledge Components (KCs) to automatically identify code-based learning resources targeting similar concepts. In our approach, pattern-based KCs are extracted from each code sample, and related activities are identified by measuring similarity between the KC sets associated with each activity. By leveraging alignment at the level of semantically important programming patterns, this method supports contextually appropriate and pedagogically useful recommendations. We evaluate our approach on an expert-organized corpus of introductory Python materials in which instructors grouped items i
arXiv:2607.05408v1 Announce Type: new Abstract: The environmental impact of training large language models (LLMs) is increasingly scrutinised, yet most published estimates focus on operational energy and disclose little about manufacturing (embodied) emissions, water consumption, or the underlying highperformance computing (HPC) infrastructure. We present a life cycle assessment (LCA) of the pre-training of Lucie 7B, an open-source multilingual Foundation Model developed by the OpenLLM-France consortium and trained on the NVIDIA H100 partition of the Jean Zay supercomputer operated by IDRIS (CNRS). The assessment is framed by the AFNOR SPEC 2314 "Frugal AI" reference and applies the Labos 1point5 methodology for greenhouse gas(GHG) accounting in computing. The study scope extends from data preparation to model validation, and integrates the full life cycle of the hardware infrastructure: manufacturing (including raw-material extraction), use (compute, temporary storage, system administ
arXiv:2607.05407v1 Announce Type: new Abstract: Modern artificial intelligence (AI) systems present profound new risks to child safety. AI is increasingly being misused to create AI-generated child sexual abuse material, facilitate child sexual exploitation, and reduce barriers to harm. In this paper, we argue that protecting children from AI-facilitated sexual abuse requires new approaches to AI safety. Existing safety techniques assume data accessibility, transparency, and evaluation practices that are incompatible with the ethical and legal constraints surrounding child sexual abuse material. We examine how these constraints create new technical challenges, such as limitations on dataset auditing, red teaming, and fine-tuning prevention. In turn, we outline *15 open problems* in online child sexual exploitation and abuse across the AI development lifecycle, from dataset curation and model design to deployment and long-term maintenance. We propose targeted recommendations for researc
arXiv:2607.05406v1 Announce Type: new Abstract: Large language models generate code from natural language prompts, enabling "vibe coding," which allows non-programmers to develop computational solutions. Vibe coding for teachers amplifies the value of teachers-as-designers, improving technology integration while fostering AI literacy. However, structured guidance on supporting this process is lacking. We propose GAIDE (A Guiding Framework for AI-Integrated Design for Educators), a framework that supports K-12 teachers in creating AI-powered learning technologies through vibe coding. The initial framework, built on Design Thinking and INTERACT, was validated through a CORDTRA interaction analysis of three teachers and four faculty mentors in an eight-week workshop to derive the final framework. Additionally, the qualitative analysis of pre- and post-interviews found an enhancement of teachers' AI literacy. Findings highlight the potential of learning-by-creating for professional develop
arXiv:2607.05405v1 Announce Type: new Abstract: To interact with users fairly and without stereotyping, AI models must display cultural competency, i.e., the ability to infer and adapt to a user's implicitly signaled cultural values, rather than relying on static demographic traits. We introduce CCBENCH, a framework for evaluating cultural competency in large language models (LLMs), treating culture as a continuum of norm adherence states rather than as a binary state of cultural belongingness. As a case study on health, we create CCBENCH-Health, which includes 60 theoretically grounded personas exhibiting varied norm-adherence states across six cultures, each engaging in 18 realistic dialogues. Each persona is evaluated on 52 authentic healthcare questions drawn from real user forums, yielding 3,120 unique interactions. Benchmarking five leading models reveals that even the best achieve culturally appropriate responses only 20-30% of the time. When explicitly prompted to focus on cult
arXiv:2607.05404v1 Announce Type: new Abstract: Frontier AI's labor-market effects matter to workers, firms, and policymakers, but current evidence generally comes from a handful of high-income economies. The capabilities of frontier AI are jagged across work tasks and national economies diverge in how they allocate human labor. We introduce a national AI exposure metric that combines occupation-level exposure scores and international employment data for 141 countries. We find that high income countries are substantially more exposed than low income countries and that Europe and Central Asia are 50 percent more exposed than Sub-Saharan Africa. We also find a gender gap: women are more exposed than men in 91 percent of countries, driven by their concentration in white-collar and sales occupations. The exceptions are countries where women's employment remains concentrated in agriculture and household enterprises. We validate our national AI exposure estimates by showing they predict nati
arXiv:2607.05403v1 Announce Type: new Abstract: This paper aims to synthesize empirical research on AI tools used to support English as a second/foreign language (EL2) learners in Arab University classrooms (AUCs) between Jan 1st 2023 and Aug 31st 2025. We utilized 3 large datasets, namely Google Scholar, Web of Science, and Scopus as the data sources. Using PRISMA-guided searches across these well-known databases, we included only published articles. The search process results in 184 studies, but only 11 studies met the inclusion criteria. Findings unveil that EL2 learners have positive attitudes towards AI for drafting, revision, and practice. Empirical gains were most consistent for surface-level outcomes improvements in higher-order writing quality and speaking proficiency was mixed and often contingent on teacher mediation. The paper concludes by proposing a research agenda and practical guidelines for Arab universities seeking evidence-based AI integration in EL2 instruction. It
arXiv:2607.05402v1 Announce Type: new Abstract: Improvements in AI technologies have made it feasible to develop new types of medical AI tools. However, these tools raise new kinds of questions, especially in relation to the ethics and AI Act compliance. We analyzed two cases of AI tools developed to predict medical risks, the risk of work disability (case A) and the risk of getting Alzheimer's disease (case B). We observed both cases using the ethical AI and the EU AI Act as frameworks, noted that they classify as high-risk systems, and that bringing them from the research environment to production would require a lot of work and compliance due to the related regulation.
When Misdirected Use of AI Broke Graduation Ceremonies
Late last year, members of Congress met to scrutinize college costs and to press institutions to be more transparent about what students pay and what they get in return. But while the hearing focused on dollars and cents, the price of college takes many forms. The post The hidden cost of college isn’t money–it’s time and opportunity appeared first on eCampus News .
Ambient artificial intelligence is being hyped as a new wave of AI with the potential to make today’s tech-enabled classrooms even smarter. Compared with the transactional nature of traditional AI, ambient AI “fades into the environment rather than sitting in a visible tool waiting for someone to type a prompt,” explains Narmeen Makhani, founder of AIxecute, a strategic advisory and consulting firm. The technology is already being employed in medical settings, helping clinicians with note-taking and after-visit summaries. To pick up on engagement and classroom interaction in real time,…
The E-Rate program, a federal initiative that has been providing broadband discounts to K–12 districts since 1998, is currently under review, and experts are asking individuals to advocate for the program during a public comment period. That was the messaging at ISTELive 26 in Orlando, Fla., where Dave LeNard, E-Rate manager for CDW, and Amy Passow, senior manager of education funding solutions for CDW, spoke to school and district leaders about the importance of saving this program. What Is E-Rate? The E-Rate program is designed to provide connectivity to schools and libraries. It…
These annual awards celebrate the groundbreaking products exhibited at ISTE that are transforming education in schools around the world.
New edtech products that have caught our attention this month
Leaders are decisive for the success of institutions and proper fit is decisive for the success of leaders. Your college doesn’t only need a good leader; you need the right leader for your organization in 2026 and beyond. The post What skills are university leaders prioritizing in new hires? appeared first on eCampus News .
arXiv:2606.19501v2 Announce Type: replace-cross Abstract: Decentralized finance exposes supervisors to fast-moving, networked credit risks. General-purpose LLM agents fit this setting poorly: they over-read weak evidence and recommend high-stakes interventions, while existing evaluations offer no regulator-aligned way to measure the resulting false alarms. We introduce DeXposure-Claw, a forecast-grounded agentic supervision system that routes LLM decisions through structured evidence: (1) DeXposure-FM, a graph time-series foundation model, forecasts future exposure networks; (2) deterministic monitors and stress scenarios then turn those forecasts into typed alerts, attribution signals, and scenario evidence; and (3) data-health and confidence gates constrain escalation before DeXposure-Claw emits auditable supervisory tickets with rationales. We further develop DeXposure-Bench, a six-axis evaluation harness, whose decision axis scores tickets against a regulator-aligned absolute-loss
arXiv:2606.14027v3 Announce Type: replace-cross Abstract: Agentic browsers integrate autonomous AI agents into web browsers, enabling users to accomplish web tasks through natural-language instructions. The same-origin policy (SOP) is a fundamental browser security mechanism that prevents unauthorized automated cross-origin data flows induced by scripts. However, whether SOP remains effective in agentic browsers is an open question that has not been systematically studied. In this work, we bridge this gap. We first observe that an agentic browser can itself serve as an automated channel for cross-origin data flows, potentially leading to SOP violations. To investigate this phenomenon, we construct SOPBench, a benchmark for evaluating SOP violations in agentic browsers. Our evaluation shows that existing agentic browsers frequently violate SOP, both in benign settings and under attacks. To address this problem, we propose SOPGuard, an SOP enforcement mechanism tailored to agentic browse
arXiv:2606.13239v2 Announce Type: replace-cross Abstract: Existing computer-use agents remain fundamentally limited in professional software manipulation: GUI-based agents suffer from fragile visual grounding and long-horizon error accumulation, while API-basedapproaches struggle with heterogeneous protocols and inaccessible commercial interfaces. In this work,we identify the Component Object Model (COM) as a unified executable abstraction, proposing COM-as-Action: a new paradigm that reframes professional software interaction as deterministic program synthesisrather than sequential visual control. To validate this paradigm in the most demanding environments, weintroduce ComCADBench, the first benchmark for agents operating real industrial CAD software. Ourexperiments reveal a substantial paradigm gap: frontier proprietary models achieve near-zero successunder GUI-based interaction, whereas COM-based execution yields substantial immediate gains. Tobridge the remaining gap between synta
arXiv:2606.12634v2 Announce Type: replace-cross Abstract: Long-horizon tool-use reinforcement learning learns from outcome verification, but trajectory-level advantages are broadcast over reasoning, API, and answer tokens. Direct self-distillation can supply a denser signal, but in our experiments it can also destroy tool use by rehearsing teacher behavior without identifying which actions the verifier rewards. We introduce Sibling-Guided Credit Distillation (SGCD), which uses distillation for bounded credit weighting rather than as a competing actor loss. Dynamic sampling produces mixed successful and failed sibling rollouts; an external LLM summarizes their contrast into a training-only credit reference; and detached teacher/student divergence reshapes GRPO token advantages. The deployed student receives only the clean task prompt. Across AppWorld and tau^3-airline, SGCD reports higher held-out point estimates than GRPO-family comparators: AppWorld TGC improves from 42.9 to 45.6 on t
arXiv:2606.09052v3 Announce Type: replace-cross Abstract: Self-evolution offers a scalable path to stronger reasoning: a pretrained language model improves itself with only minimal external supervision. Yet existing methods either depend on extensively curated or teacher-generated training data, or, when the generator runs unsupervised, reward it by a difficulty heuristic that need not improve the solver. We introduce INFUSER, an iterative co-training framework with two co-evolving roles: a Generator that drafts questions and reference golden answers from a pool of unstructured, automatically collected documents, and a Solver that improves by training on them. The solver is trained with standard correctness rewards against the generator-provided answers, while the generator is rewarded by an optimizer-aware influence score that measures whether each proposed question would actually improve the solver on the target distribution. Because this continuous, noisy influence score is poorly s
arXiv:2605.24661v2 Announce Type: replace-cross Abstract: LLMs have achieved remarkable success in complex reasoning tasks, yet current evaluation approaches predominantly rely on final-answer correctness, offering limited insight into the underlying reasoning processes that produce those answers. To address this gap, this study proposes a unified multi-dimensional framework for measuring reasoning quality in LLMs from a behavioral perspective, operationalizing six theoretically grounded dimensions: Correctness (CQ), Consistency (CS), Robustness (RS), Logical Coherence (LS), Efficiency (ES), and Stability (SS). Extensive experiments on seven LLMs across 975 items from four benchmarks demonstrate that the framework reveals behaviors invisible to accuracy-only metrics. Notably, logical coherence is orthogonal to correctness (r = -0.172, ns), confirming that correct answers can arise from incoherent reasoning, while Claude-Haiku-4.5 achieves the highest multi-dimensional score (Q_bal = 0.
arXiv:2605.09165v2 Announce Type: replace-cross Abstract: Looped language models repeat a set of transformer layers through depth, reducing memory costs and providing natural early-exit points at loop boundaries. However, looped models do not scale as favorably as standard transformers with unique layers. We compare standard and Mixture-of-Experts (MoE) transformers, with and without looping, and find two main results. First, we find Looped-MoE models scale better than the standard baseline while dense looped models do not. We trace this to routing divergence between loops: in Looped-MoE models, different experts are activated on each pass through the same shared layers, recovering expressivity without additional parameters. Our second finding is that looped models have better compute-quality trade-offs with early exits than standard models. Because each loop ends with the same layers that produce the final output, loop boundaries are superior exit points, as confirmed by earlier outpu
arXiv:2604.21495v2 Announce Type: replace-cross Abstract: Numerical reasoning over expert-domain tables often exhibits high in-domain accuracy but limited robustness to domain shift. Models trained with supervised fine-tuning (SFT) on specific datasets tend to rely on header-operation shortcuts rather than structural reasoning. We introduce TaNOS, a continual pre-training framework comprising three components: (i) header anonymization to reduce lexical memorization, (ii) operation sketches that provide minimal structural cues, and (iii) self-supervised pretraining that constructs correctness-guaranteed program-question pairs from given tables in a program-first manner. By decoupling domain semantics and numerical operation structure, TaNOS improves the transferability of numerical reasoning. Applied to an 8B instruction-tuned model, TaNOS achieves 80.13% execution accuracy on FinQA with only 10% train data, outperforming SFT baseline (73.97%) with full train data and proprietary models
arXiv:2603.14732v2 Announce Type: replace-cross Abstract: As large language models (LLMs) are increasingly considered for automated assessment and feedback, understanding when LLM marking is valid is essential. We evaluate LLM-as-a-judge marking across three physics assessment formats - structured questions, written essays, and scientific plots - comparing GPT-5.2, Grok 4.1, Claude Opus 4.5, DeepSeek-V3.2, Gemini Pro 3, and committee aggregations against human markers under blind, solution-provided, false-solution, and anchored conditions. We distinguish absolute accuracy from rank-order agreement, since a marking system can match the distribution of human marks while failing to order responses by quality. Across task types, performance is sharply task-dependent. For blind university exam questions ($n=771$) and secondary and university structured questions ($n=1151$), models show robust rank-order agreement with human markers (Spearman $\rho > 0.6$), with official solutions reducing e
arXiv:2602.15029v3 Announce Type: replace-cross Abstract: The internal representations learned by language models consistently exhibit striking geometric structure: calendar months organize into a circle, historical years form a smooth one-dimensional manifold, and cities' latitudes and longitudes can be decoded using a linear probe. To explain this neural code, we first show that language statistics exhibit translation symmetry (for example, the frequency with which any two months co-occur in text depends only on the time interval between them). We prove that this symmetry governs these geometric structures in high-dimensional word embedding models, and we analytically derive the manifold geometry of word representations. These predictions empirically match large text embedding models and large language models. Moreover, the representational geometry persists at moderate embedding dimension even when the relevant statistics are perturbed (e.g., by removing all sentences in which two m
arXiv:2602.11354v3 Announce Type: replace-cross Abstract: The literature has witnessed an emerging interest in AI agents for automated assessment of scientific papers. Existing benchmarks focus primarily on the computational aspect of this task, testing agents' ability to reproduce or replicate research outcomes when having access to the code and data. This setting, while foundational, (1) fails to capture the inconsistent availability of new data for replication as opposed to reproduction, and (2) lacks ground-truth diversity by focusing only on reproducible papers, thereby failing to evaluate an agent's ability to identify non-replicable research. Furthermore, most benchmarks only evaluate outcomes rather than the replication process. In response, we introduce ReplicatorBench, an end-to-end benchmark, including human-verified replicable and non-replicable research claims in social and behavioral sciences for evaluating AI agents in research replication across three stages: (1) extrac
arXiv:2601.18778v3 Announce Type: replace-cross Abstract: RL methods for scaling large reasoning models stall on datasets with low initial success rates, and thus little training signal. We investigate a fundamental question: Can a pretrained LLM leverage latent knowledge to generate an automated curriculum for problems it cannot solve? We explore this with SOAR: An asymmetric self-play framework that uses meta-RL to surface these pedagogical signals. A teacher model proposes synthetic problems for a student model, and is rewarded with its improvement on a subset of hard problems, thus grounding the curriculum in real student progress rather than intrinsic proxy rewards. Our study on the hardest subsets of math benchmarks (0/128 success) reveals three core findings. First, it is possible to realize bilevel meta-RL that unlocks learning under sparse, binary rewards by sharpening a latent capacity of pretrained models to generate useful problems. Second, grounded rewards outperform intri
arXiv:2601.13932v2 Announce Type: replace-cross Abstract: Reaching consensus on massive discussion networks is critical for reducing noise and achieving optimal collective outcomes. However, the natural tendency of humans to preserve their initial ideas constrains the emergence of global solutions. To address this, Collective Intelligence (CI) platforms facilitate the discovery of globally superior solutions. We introduce a dynamical system based on the standard $O(N)$ model to drive the aggregation of semantically similar ideas. The system consists of users represented as nodes in a $d=2$ lattice with nearest-neighbor interactions, where their ideas are represented by semantic vectors computed with a pretrained embedding model. We analyze the system's equilibrium states as a function of the coupling parameter $\beta$. Our results show that $\beta > 0$ drives the system toward a ferromagnetic-like phase (global consensus), while $\beta < 0$ induces an antiferromagnetic-like state (maxi
arXiv:2510.19600v2 Announce Type: replace-cross Abstract: In the quest for scientific progress, communicating research is as vital as the discovery itself. Yet, researchers are often sidetracked by the manual, repetitive chore of building project webpages to make their dense papers accessible. While automation has tackled static slides and posters, the dynamic, interactive nature of webpages has remained an unaddressed challenge. To bridge this gap, we reframe the problem, arguing that the solution lies not in a single command, but in a collaborative, hierarchical process. We introduce $\textbf{AutoPage}$, a novel multi-agent system that embodies this philosophy. AutoPage deconstructs paper-to-page creation into a coarse-to-fine pipeline from narrative planning to multimodal content generation and interactive rendering. To combat AI hallucination, dedicated "Checker" agents verify each step against the source paper, while optional human checkpoints ensure the final product aligns perfe
arXiv:2510.09340v2 Announce Type: replace-cross Abstract: Recent language models exhibit significant logical reasoning abilities, yet the mechanisms supporting deductive inference remain poorly understood. This paper studies small transformer-based language models trained from scratch on multi-step deductive tasks, focusing on the distinction between horizontal reasoning, where intermediate steps are generated autoregressively, and vertical reasoning, where inference unfolds implicitly across layers before the first output token is produced. We analyze two synthetic tasks: logical consequence over chains of symbolic implications and root-to-leaf navigation in binary trees. Mechanistic interpretability reveals that Chain-of-Thought supervision enables models to learn rule-based inference rather than statistical shortcuts. In the horizontal setting, a shallow attention-only model develops interpretable circuits for rule completion, rule chaining, and final decision making, largely implem
arXiv:2606.26493v2 Announce Type: replace Abstract: Diffusion language models offer a promising alternative to autoregressive models due to their potential for parallel and iterative generation. However, existing approaches use a single network for both context representation and iterative denoising, forcing one model to serve both roles and limiting its capacity for either role. We propose TwoTower, a block-wise autoregressive diffusion model that decouples these roles into two towers: a frozen AR context tower that causally processes clean tokens, and a trainable diffusion denoiser tower with bidirectional block attention that refines noisy blocks via cross-attention to the context. Built on Nemotron-3-Nano-30B-A3B, an open-weight 30B hybrid Mamba-Transformer MoE model, and trained on approximately 2.1T tokens, Nemotron-Labs-TwoTower retains 98.7% of the autoregressive baseline's quality while offering 2.42X higher wall-clock generation throughput. We release the code and model weigh
arXiv:2606.23375v2 Announce Type: replace Abstract: While the wider applicability of LLMs in the legal field is currently debated due to their reliability and the gravity of any errors, narrow uses with well-understood and mitigated risks have emerged. Notably the Swiss Federal Supreme Court uses small on-premises models for tentative translations and short-passage summarization across the four official languages. However, such usage is challenging in the context of Criminal Law. Since rulings and cases employees work on routinely can contain detailed descriptions of violent and sexual offenses, their legitimate work is compromised by refusals and disclaimers due to the activation of model guardrails (over-alignment). To measure this phenomenon, we introduce TF-RefusalBench, a multilingual benchmark for criminal-law translation and summarization derived from public Swiss Supreme Court rulings. TF-RefusalBench contains 5,200 total prompts across French, German, Italian, and English, cor
arXiv:2606.22723v2 Announce Type: replace Abstract: Although Large Language Models (LLMs) excel in many tasks, their assessment in Portuguese has received less attention, particularly for open-ended, discursive tasks that demand deeper reasoning and generation capabilities. While the original BLUEX benchmark addressed the scarcity of Portuguese evaluation datasets through multiple-choice questions from Brazilian university entrance exams, it did not cover the more challenging second-phase examinations, which require free-form written responses. In this work, we introduce BLUEX v2, a benchmark derived from the second-phase entrance exams of Brazil's two leading universities: UNICAMP (Comvest) and USP (Fuvest), spanning exam years 2022--2025. Our dataset comprises 395 questions unfolding into 919 graded subquestions, with 55.7% of questions containing associated images (represented as context-aware captions during inference to enable evaluation across both vision-capable and text-only mo
arXiv:2606.05895v2 Announce Type: replace Abstract: Research metrics increasingly use attention as evidence of societal impact. Yet attention serves as evidence only once interpreted, and its meaning depends on the contexts in which it occurs, not on volume alone. Altmetrics records signals in isolation, retaining a count of the attention an output received, or a sequence of when. We address this gap with attention flows, representations that situate a research output's attention in the social settings in which it occurs, the language expressing it, and the time over which it arrives. To evaluate the flow, we construct a benchmark of analogy queries, each testing whether the relationship between two outputs transfers to a third. The count and sequence baselines fail to recover these relationships, whereas flows learned with dynamic contextualised embeddings recover them. The recovered structure survives partial observation and is intrinsic to the attention itself. These findings suppor
arXiv:2605.28183v3 Announce Type: replace Abstract: We introduce BenGER (Benchmark for German Law), a benchmark and dataset for evaluating LLM systems on subsumption-based legal reasoning in German law. The dataset combines 596 exam-style free-text legal case tasks across multiple levels of legal education and 531 short doctrinal reasoning tasks. It includes a controlled validation subset of timed human-written solutions under both unaided and human-AI co-creation conditions. We evaluate 12 contemporary LLM systems - closed flagship, efficiency-oriented, and open-weight - with a rubric-aligned LLM-as-a-Judge cross-validated against a multi-rater human-grading layer (three blind reviews per solution, six judge families benchmarked against the human pool). Closed-flagship systems lead the leaderboard across all three corpora, human-AI co-creation measurably improves on unaided human work, and the LLM judge tracks human grading at Pearson r=0.76 and Cohen's \k{appa}=0.60. System rankings