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.
Pomona College, a member of The Claremont Colleges located in Claremont, Calif., is a longtime leader in campus sustainability policies and is working to become carbon-neutral by 2030. Those efforts extend into many facets of campus life, including the college’s self-operated dining services, where they meet head-on with another Pomona College goal: technological innovation. Pomona College’s Dining Services staff is using an artificial intelligence-driven solution from Metafoodx in the dining hall kitchens that prepare meals for the student population of 1,700 to support campuswide…
New York State is betting big on Portrait of a Graduate. It could transform how schools think about student success — or become one of the most well-intentioned failures in recent education policy. Last summer, the state formally adopted its Portrait framework, committing to a new vision of what it means to earn a diploma […]
The Financial Aid Community Needs Clarity and Accountability From ED sara.custer@in… Wed, 08/19/2026 - 10:20 AM And it needs more critical coverage of the sweeping regulatory changes administrators are being asked to implement. Byline(s) Letters to the Editor
The next healthcare revolution isn’t about treating osteoporosis, it’s about preserving mobility and healthspan. The post Why Bone Health Is Preventive Healthcare’s Next Big Category appeared first on MedCity News .
The Greek epic’s box office success can provide an inroad for students to understand how people have used myth to explain the world around them.
Educators should also be wary of artificial intelligence when teaching students to write high-stakes items like college essays or cover letters, experts say.
This summer, I ran a one-week artificial intelligence pilot at the Chicago headquarters of Forging Opportunities for Refugees in America, a nonprofit dedicated to helping refugees succeed in school. Most of the students were still building their literacy skills, and several of the 22 newcomers, ranging in age from 11 to 17, had arrived from […]
A school in Kentucky sent students home with a bound agenda that included absurd errors clearly made by AI. The periodic table was nonsense. The map of the U.S. lists the state of Illinois as “Vitoiis,” and Kentucky is listed as “Venecky.” The post Which AI chatbot did this school use to make such a ridiculous map? appeared first on District Administration .
There is broad consensus among psychologists that corporal punishment can be harmful to children, but the U.S. has no federal laws banning it in public schools. Instead, it's up to states and local districts to decide. The post 15 states have laws that allow corporal punishment in schools. Here’s where, and why appeared first on District Administration .
Under state standards, metrics like adult-to-child ratios and accessibility would worsen, according to a Children’s Equity Project analysis.
Indiana has more than 500 job openings for medical assistants, but training for the job isn’t cheap — between $1,500 and $4,000 at lower-cost schools, such as Indiana’s Ivy Tech Community College. The reduced $1,500 tuition for low-income people the college offers is still a challenge for students trying to bootstrap their way to an […]
If you ask most educators where college and career readiness (CCR) begins, many will point to high school. That's when many students start thinking seriously about graduation requirements, dual enrollment, financial aid, career pathways, and college applications.
These mindfulness apps are ideal for use in schools to help students better focus and improve their overall wellbeing
AI-generated content is not automatically plagiarism. That statement may unsettle faculty members and academic-integrity officers. It may sound like a defense of students who use ChatGPT to write papers or complete assignments intended to demonstrate mastery. The post Stop calling every misuse of AI plagiarism appeared first on eCampus News .
Jacksonville State Dissolves Faculty Senate Emma Whitford Wed, 08/19/2026 - 03:00 AM Byline(s) Emma Whitford
Focus on the Quantum Future jdimaggio@upcea.edu Wed, 08/19/2026 - 03:00 AM We seem to be obsessed with artificial general intelligence and super intelligence. Yet, our forward focus today more appropriately should be on quantum computing. Byline(s) Ray Schroeder
Universities of Wisconsin Announce 2-Year Tuition Freeze Emma Whitford Wed, 08/19/2026 - 03:00 AM Byline(s) Emma Whitford
Lawsuit Challenges New Length-of-Stay Limit for International Students Johanna Alonso Wed, 08/19/2026 - 03:00 AM The rule, finalized last month, has drawn intense criticism from universities and international educators. Now, a group of eight organizations is arguing that it should be scrapped. Byline(s) Johanna Alonso
AI Career Skills Move Beyond Computer Science Joshua.Bay Wed, 08/19/2026 - 03:00 AM New Handshake data shows how students across majors are building AI experience and navigating an evolving early-career job market. Byline(s) Joshua Bay
Florida Responds to McMahon ‘Call to Action’ Josh Moody Wed, 08/19/2026 - 03:00 AM Byline(s) Josh Moody
Calif. Bill Could Create Designation for Asian-Serving Colleges Olivia.sanchez Wed, 08/19/2026 - 03:00 AM The proposal doesn’t have a funding component but would require institutions to establish plans to better serve Asian American, Native Hawaiian and Pacific Islander students. Byline(s) Olivia Sanchez
Students’ Choices of Roommates, Clubs and More Fuel Higher Ed Polarization Ryan Quinn Wed, 08/19/2026 - 03:00 AM A new book by sociologist Neil Gross argues universities aren’t doing enough to counteract students’ formation of their own ideological bubbles. Byline(s) Ryan Quinn
42% of Colleges Haven’t Fully Submitted Required Student Outcomes Data jessica.blake@… Wed, 08/19/2026 - 03:00 AM The Education Department is giving the more than 1,900 colleges missing some or all Financial Value Transparency data one last chance. Noncompliance could bring consequences. Byline(s) Jessica Blake
The National Association of School Nurses says it's "deeply concerned" with the Trump administration's goal to reform the childhood vaccine schedule.
High school came with more responsibility and less accountability for Josiah Ulukita. Courses were much harder, teachers were less strict about attendance, and it was easier to skip school with friends or go on his phone during class, said 16-year-old Ulukita, who attends Kaimuki High School in Honolulu. He went from easily passing middle school […] The post A national divide over holding back students: Some states do it young, others wait until high school appeared first on The Hechinger Report .
On This Week with EdSurge, searching to see where artificial intelligence quietly adds up.
Instead of banning the technology, schools have a unique opportunity to promote digital citizenship and authentic student engagement.
arXiv:2605.17231v2 Announce Type: replace-cross Abstract: Activation steering has emerged as a lightweight approach for modifying language model behavior without parameter updates, yet existing methods remain brittle: unstable across layers and prone to disturbing behavior unrelated to the target concept. We trace these failures to a hidden assumption shared by widely-used methods such as CAA, ActAdd, and ITI: that the intermediate activation space is Euclidean. We show this assumption is fundamentally flawed. The metric that actually governs how a hidden-state perturbation changes the output is the Fisher information metric of the softmax layer, pulled back to the intermediate layer through the Jacobian of the intervening layers. From it we derive a closed-form steering direction, applied to a hidden state at an intermediate layer, that reaches a target concept change with the least non-target distortion. The framework is sharpest in the early and middle intermediate layers, where the
arXiv:2604.15794v2 Announce Type: replace-cross Abstract: Large Language Models (LLMs) have achieved remarkable success, underpinning diverse AI applications. However, they often suffer from performance degradation due to factors such as catastrophic forgetting during Supervised Fine-Tuning (SFT), quantization, and pruning. In this work, we introduce a performance recovery framework based on Self-Distillation Fine-Tuning (SDFT) that effectively restores model capabilities. Complementing this practical contribution, we provide a rigorous theoretical explanation for the underlying recovery mechanism. We posit that an LLM's generative capability fundamentally relies on the high-dimensional manifold constructed by its hidden layers. To investigate this, we employ Centered Kernel Alignment (CKA) to quantify the alignment between student and teacher activation trajectories, leveraging its invariance to orthogonal transformations and scaling. Our experiments demonstrate a strong correlation b
arXiv:2604.02118v2 Announce Type: replace-cross Abstract: Natural language explanations of time series data are increasingly produced by foundation models in high stakes domains, making factual correctness critical. Evaluating such explanations differs fundamentally from standard natural language generation: correctness requires verifying numerical claims against structured data rather than similarity to reference text. While LLM as a Judge has emerged as a scalable paradigm for text evaluation, its applicability to numerically grounded time series explanations remains unstudied. We introduce TSQueryBench, a controlled synthetic benchmark of 500 time series instances across 10 query types, each paired with correct, partially correct, and incorrect explanations. We evaluate six large language models across four tasks: explanation generation, relative ranking, independent scoring, and multi anomaly detection. Our central finding is a consistent generation evaluation asymmetry: models tha
arXiv:2606.21802v2 Announce Type: replace Abstract: Standard tokenwise diffusion LMs keep training corruption and inference commitment at token granularity throughout denoising. At high noise, this leaves scattered local fragments rather than coherent evidence, making it hard to form early coarse structure, exactly what planning-sensitive generation requires. Hierarchical planning methods add coarse stages to separate planning from wording, but they need extra planners, block latents, or two stage designs. We propose Noise Dependent Granularity Control (NDGC), a single-level diffusion method that uses the noise level as a granularity cue. NDGC aligns training exposure and inference commitment with denoising progress. High noise steps use coherent token groups to support early meaning commitment, while low noise steps return to token level refinement. This creates planning like coarse to fine denoising without an explicit planner or hierarchical architecture. Across controlled tests, ab
arXiv:2605.31281v2 Announce Type: replace Abstract: As wind turbine fleets age, data-driven reliability engineering and maintenance optimisation are essential to manage lifecycle expenditure and support asset life extension. Historical maintenance records offer a vital source of field evidence, yet their analytical use is impeded by inconsistent system codes, generic categorical fields, and unstructured technician text. This paper presents a topology-aware large language model (LLM) workflow for reviewing legacy labels, extracting candidate maintenance and failure-mode taxonomies, and assigning structured semantic fields at record level. The workflow processed 16,316 maintenance records from 280 turbines across 32 onshore wind farms, spanning 9.2 years of operational history. It combines system-specific batch synthesis with granular labelling, deterministic exclusions, structured outputs, record-level provenance, and explicit review routes. Of 2,984 records targeted by three system-cod
arXiv:2605.24930v2 Announce Type: replace Abstract: Transformer-based LLMs achieve strong results on many language tasks; however, long inputs remain challenging because context windows are finite, and prefill latency and memory grow rapidly with prompt length. Flat token-stream processing and chunk-based retrieval can therefore spend substantial computation and context budget on text unrelated to the query. Offline-indexed RAG additionally introduces external storage and index management overhead, and typically appends retrieved evidence as raw text, increasing prefill cost and latency. H^{2}MT makes long-context inference structure-aware: it builds a semantic hierarchy offline, computes a memory embedding for each node via bottom-up post-order aggregation, and routes queries coarse-to-fine at inference to prune irrelevant branches early. On LongBench QA (NarrativeQA, HotpotQA, QASPER) and two structured technical-document settings, H MT achieves favorable quality efficiency trade-off
arXiv:2605.17443v3 Announce Type: replace Abstract: We analyze how automatic speech recognition (ASR) errors propagate through ASR--LLM cascades in Korean spoken question answering (SQA), focusing on downstream semantic failures that conventional ASR metrics cannot fully capture. Our analysis shows that the relative downstream degradation caused by ASR errors is consistent across LLMs with different absolute performance, suggesting that cascade degradation largely tracks ASR-stage information loss. We further identify single-character ASR errors as a particularly salient source of information loss in Korean, where even a minimal transcription difference can change the intended question and degrade downstream QA performance. Finally, an auxiliary comparison shows that a large audio language model outperforms an ASR--LLM cascade with an approximately matched language backbone in noisy Korean SQA, indicating the potential of direct audio input to mitigate transcript-induced information lo
arXiv:2605.07507v2 Announce Type: replace Abstract: The rapid growth of academic publications has created a need for tools that extract structured knowledge from unstructured scientific texts. Although large language models (LLMs) can perform natural language understanding and information extraction, existing solutions often require specialized infrastructure, programming expertise, or fine-tuned domain-specific models, which limits their accessibility for researchers in specialized fields. This paper describes TCMIIES (Traditional Chinese Medicine Information Intelligent Extraction System), a browser-based, zero-installation platform that uses commercial LLM APIs to perform structured information extraction from academic literature. The system employs a schema-guided prompting framework with automatic system prompt generation, allowing researchers to define custom extraction schemas through a graphical interface without programming. TCMIIES features a pure front-end architecture that
arXiv:2604.20817v2 Announce Type: replace Abstract: Language models trained on natural text learn to represent numbers using periodic features with dominant periods at $T=2, 5, 10$. In this paper, we identify a two-tiered hierarchy of these features: while Transformers, Linear RNNs, LSTMs, and classical word embeddings trained in different ways all learn features that have period-$T$ spikes in the Fourier domain, only some learn geometrically separable features that can be used to linearly classify a number mod-$T$. To explain this incongruity, we prove that Fourier domain sparsity is necessary but not sufficient for mod-$T$ geometric separability. Empirically, we investigate when model training yields geometrically separable features, finding that the data, architecture, optimizer, and tokenizer all play key roles. In particular, we identify two different routes through which models can acquire geometrically separable features: they can learn them from complementary co-occurrence sign
arXiv:2604.06474v2 Announce Type: replace Abstract: Deep research with Large Language Model (LLM) agents is emerging as a powerful paradigm for multi-step information discovery, synthesis, and analysis. However, existing approaches primarily focus on unstructured web data, while the challenges of conducting deep research over large-scale structured databases remain relatively underexplored. Unlike web-based research, effective data-centric research requires more than retrieval and summarization and demands iterative hypothesis generation, quantitative reasoning over structured schemas, and convergence toward a coherent analytical narrative. In this paper, we present DataSTORM, an LLM-based agentic system capable of autonomously conducting research across both large-scale structured databases and internet sources. Grounded in principles from Exploratory Data Analysis and Data Storytelling, DataSTORM reframes deep research over structured data as a thesis-driven analytical process: disco
arXiv:2604.06416v2 Announce Type: replace Abstract: Although LLM context lengths have grown, there is evidence that their ability to integrate information across long-form texts has not kept pace. We evaluate one such understanding task: generating summaries of novels. When human authors of summaries compress a story, they reveal what they consider narratively important. Therefore, by comparing human and LLM-authored summaries, we can assess whether models mirror human patterns of conceptual engagement with texts. To measure conceptual engagement, we align sentences from 150 human-written novel summaries with the specific chapters they reference. We demonstrate the difficulty of this alignment task, which indicates the complexity of summarization as a task. We then generate and align additional summaries by nine state-of-the-art LLMs for each of the 150 reference texts. Comparing the human and model-authored summaries, we find both stylistic differences between the texts and difference
arXiv:2603.24472v4 Announce Type: replace Abstract: Self-distillation has emerged as an effective post-training paradigm for LLMs, often improving performance while shortening reasoning traces. However, in mathematical reasoning, we find that it can reduce response length while degrading performance. We trace this degradation to the suppression of epistemic verbalization - the model's expression of uncertainty during reasoning. Through controlled experiments varying conditioning context richness and task coverage, we show that conditioning the teacher on rich information suppresses uncertainty expression, enabling rapid in-domain optimization with limited task coverage but harming OOD performance, where unseen problems benefit from expressing uncertainty and adjusting accordingly. Across Qwen3-1.7B/8B, DeepSeek-Distill-Qwen-7B, and Olmo3-7B-Instruct, we observe performance drops of up to 40%. Our findings highlight that exposing appropriate levels of uncertainty is crucial for robust r
arXiv:2603.23047v2 Announce Type: replace Abstract: Retrieval-Augmented Generation (RAG) fine-tuning has shown substantial improvements over vanilla RAG, yet most studies target document question answering, leaving open whether these gains transfer to specialized tasks. We study supervised RAG fine-tuning (RAG-SFT) for requirements document generation in the electronics engineering domain, adapting two 7B models under two different training data strategies. Because Rouge and BertScore poorly capture factuality on long technical text, we introduce C-FEX, a claim-based evaluation pipeline that attributes each response claim to its origin (augmented prompt or reference response), and propose Parametric Knowledge Precision (PKP), which isolates claims originating from the model's weights and measures their correctness. We show that a prior metric to assess parametric knowledge decomposes as PKP $\times$ PR, separating the rate of parametric output (PR) from its quality (PKP). Empirically,
arXiv:2603.09872v2 Announce Type: replace Abstract: Recent work has found that contemporary language models such as transformers can become so good at next-word prediction that the probabilities they calculate become worse for predicting naturalistic reading time. In this paper, we propose that this can be explained by reading time being shaped by simple n-gram statistics rather than the more complex statistics learned by state-of-the-art transformer language models. We demonstrate that the neural language models whose predictions are most correlated with n-gram probability are also those that calculate probabilities that are the most correlated with eye-tracking-based metrics of reading time on naturalistic text.
arXiv:2603.06505v2 Announce Type: replace Abstract: Automatic speech recognition (ASR) has benefited from advances in pretrained speech and language models, yet most systems remain constrained to monolingual settings and short, isolated utterances. While recent efforts in context-aware ASR show promise, two key challenges persist: limited multilingual support and the absence of principled alignment between speech and contextual representations. In this paper, we introduce a context-aware multilingual ASR framework that supports diverse languages and accents while preserving the modularity of pretrained models. Our approach combines a frozen speech encoder and a decoder-only language model via a lightweight projection module, allowing structured context prompts, including dialogue history and biasing words, to guide transcription. To improve interaction between speech and context, we employ a contrastive learning objective that aligns their representations in a shared embedding space. E
arXiv:2603.02876v2 Announce Type: replace Abstract: Large Language Model personas, explicit profiles specifying a user's attributes, preferences, and behavioural tendencies, are increasingly used to simulate human conversations for user modelling, social reasoning, and behavioural analysis. Evaluating whether such simulations faithfully reflect human conversational behaviour is critical, yet current practice often relies on LLM-as-a-judge approaches that provide limited grounding in observable behaviour and produce opaque scalar scores. We present Eval4Sim, an evaluation framework that measures alignment between simulated and human conversations across three dimensions: adherence, whether persona traits are recoverable from dialogue via dense retrieval; consistency, whether a persona maintains a distinguishable stylistic identity via authorship verification; and naturalness, whether conversations exhibit human-like turn-to-turn flow via dialogue NLI. Unlike optimization-oriented metric
arXiv:2602.17127v2 Announce Type: replace Abstract: Large language models increasingly serve as reasoning layers in multi-agent systems, where one provider's models may generate, judge, and summarize within a single pipeline. This raises the question of whether developer organizations impart durable behavioral tendencies that could compound across such stacks. We apply a scenario-based forced-choice instrument to 18 governance-relevant behavioral dimensions across 18 models from six developer organizations. Items are model-generated, filtered by independent judges, and administered with probe blanks embedded among semantically orthogonal decoys under deterministic option shuffling. Findings are declared on effect size, requiring both Holm-corrected significance and |d| >= 0.2. Across the 14 dimensions on which one scale pole denotes a defined response failure -- sycophancy, false balance, overconfidence, and others -- organizations occupy consistent relative positions (Kendall's W = 0.
arXiv:2602.13840v2 Announce Type: replace Abstract: Large language model (LLM) agents are increasingly deployed in personalized tasks involving sensitive, context-dependent information, where privacy violations may arise in agents' action due to the implicitness of contextual privacy. Existing approaches rely on external, inference-time interventions which are brittle, scenario-specific, and may expand the privacy attack surface. We propose PrivAct, a contextual privacy-aware multi-agent learning framework that internalizes contextual privacy preservation directly into models' generation behavior for privacy-compliant agentic actions. By embedding privacy preferences into each agent, PrivAct enhances system-wide contextual integrity while achieving a more favorable privacy-helpfulness tradeoff. Experiments across multiple LLM backbones and benchmarks demonstrate consistent improvements in contextual privacy preservation, reducing leakage rates by up to 12.32% while maintaining comparab
arXiv:2602.13110v4 Announce Type: replace Abstract: Large language models (LLMs) are increasingly used as scalable judges in pairwise evaluation, but they remain prone to miscalibration and biases. We propose \textsc{Scope} (Selective Conformal Optimized Pairwise Evaluation), a framework that calibrates an acceptance threshold so that, under exchangeability, the error rate among non-abstained judgments is at most a user-specified level $\alpha$. To supply \textsc{Scope} with a bias-neutral uncertainty signal, we introduce Bidirectional Preference Entropy (BPE), which queries the judge under both response positions and converts the order-averaged preference probability into an entropy-based score. Across various pairwise judging benchmarks, BPE outperforms standard confidence proxies in calibration and discrimination, while \textsc{Scope} consistently satisfies the target risk bound (empirical FDR $\approx 0.097$--$0.099$ at $\alpha=0.10$) and retains substantial coverage. Compared to v
arXiv:2601.18899v3 Announce Type: replace Abstract: Large Language Model (LLM)-powered Automatic Speech Recognition (ASR) systems achieve strong performance with limited resources by linking a frozen speech encoder to a pretrained LLM via a lightweight connector. Prior work trains a separate connector per language, overlooking linguistic relatedness. We propose an efficient and novel connector-sharing strategy based on linguistic family membership, enabling one connector per family, and empirically validate its effectiveness across two multilingual LLMs and two real-world corpora spanning curated and crowd-sourced speech. Our results show that family-based connectors reduce parameter count while improving generalization across domains, offering a practical and scalable strategy for multilingual ASR deployment.
arXiv:2601.03570v2 Announce Type: replace Abstract: Human beings primarily understand the world through concepts (e.g., dog), abstract mental representations that structure perception, reasoning, and learning. However, how large language models (LLMs) acquire, retain, and forget such concepts during continual pretraining remains poorly understood. In this work, we study how individual concepts are acquired and forgotten, as well as how multiple concepts interact through interference and synergy. We link these behavioral dynamics to LLMs' internal concept circuits, computational subgraphs associated with specific concepts, and incorporate graph metrics to characterize circuit topology. Our analysis reveals: (1) LLMs concept circuits provide a non-trivial, consistent signal of concept learning and forgetting; (2) concept circuits exhibit a stage-wise temporal pattern during continual pretraining, with an early increase followed by gradual decrease and stabilization; (3) concepts with lar
arXiv:2511.23119v3 Announce Type: replace Abstract: High-quality main content extraction from web pages is a critical prerequisite for constructing large-scale training corpora. While traditional heuristic extractors are efficient, they lack the semantic reasoning required to handle the structural heterogeneity of the modern web. Conversely, well-pretrained generative Large Language Models (LLMs) offer superior document comprehension but are prohibited by excessive computational costs, limited context windows, and hallucination risks when applied at web scale. We present \textbf{Dripper}, a lightweight framework that resolves these bottlenecks through four contributions: (1) We reformulate extraction as a \textbf{constrained sequence labeling} task using SLMs (Small Language Models). This paradigm eliminates generative hallucinations and achieves exceptional efficiency, reaching a throughput of 3.08 pages per second on a single A100 GPU. (2) We construct \textbf{WebMainBench}, a rigoro
arXiv:2510.01832v2 Announce Type: replace Abstract: Semi-structured content in HTML tables, lists, and infoboxes accounts for a substantial share of factual data on the web, yet the formatting complicates usage, and reliably extracting structured information from them remains challenging. Existing methods either lack generalization or are resource-intensive due to per-page LLM inference. In this paper, we introduce SCRIBES (SCRIpt-Based Semi-Structured Content Extraction at Web-Scale), a novel reinforcement learning framework that leverages layout similarity across webpages within the same site as a reward signal. Instead of processing each page individually, SCRIBES generates reusable extraction scripts that can be applied to groups of structurally similar webpages. Our approach further improves by iteratively training on synthetic annotations from in-the-wild CommonCrawl data. Experiments show that our approach outperforms strong baselines by over 13% in script quality and boosts dow