USE CASEReal-World Evidence & Scientific Communication Intelligence Real-World Evidence...

USE CASEClient Portfolio Intelligence & Advisor ProductivityClient portfolio reviews...

Discussion | Link

Discussion | Link

Song Jung-a / Financial Times: Fintech app Toss, used by ~66% of South Korea's 51M people, says its FacePay service, available at 330K retail outlets, has attracted 4.8M users since September  —  Toss aims to ‘eliminate physical credit cards’ in South Korea in three years  —  A South Korean fintech company is betting payment …

Discussion | Link

Reuters: Sources: SK Hynix's big customers have been proposing a range of offers to secure memory chip supplies, including investing in dedicated chip production lines  —  SK Hynix (000660.KS) is being aggressively courted by big global tech firms with offers to invest in its new production lines …

Brian Krebs / Krebs on Security: Instructure disables its Canvas edtech platform, used by thousands of schools and universities, amid a data extortion attack claimed by ShinyHunters  —  An ongoing data extortion attack targeting the widely-used education technology platform Canvas disrupted classes and coursework …

Discussion | Link

Sam Nussey / Reuters: Sony reports 1.5M PS5 unit sales in Q4, down 46% YoY, and forecasts annual sales at its gaming business to fall 6% YoY to $1.69B, amid memory chip price surge  —  Sony (6758.T) forecast on Friday a higher annual operating profit but lower sales in its gaming business, with overall profit expected …

Discussion | Link

Discussion | Link

Like many AI companies automating work that humans currently do, Basata will eventually face a harder question about where the line is between augmenting workers and displacing them. For now, the founders say the administrative staff they work with aren't worried about that; they're more worried about drowning.

Financial Times: Sources: Anthropic weighs raising as much as $50B at a ~$900B pre-money valuation, likely closing within two months; its annualized revenue is nearing $45B  —  Start-up behind Claude tool is fielding inbound investment offers that could lead to it surpassing rival OpenAI in value

arXiv:2605.05218v1 Announce Type: new Abstract: Predictive multiplicity and chaotic dynamics represent two fundamental challenges in machine learning that have evolved independently despite their conceptual connections. We bridge this gap by introducing horizon-constrained Rashomon sets, a theoretical framework that characterizes how model multiplicity evolves with prediction horizon in chaotic systems. Unlike static prediction tasks where the Rashomon set remains fixed, chaos induces exponential divergence among initially similar models, fundamentally transforming the nature of predictive equivalence. We prove that the effective Rashomon set contracts exponentially with lead time at a rate determined by the maximum Lyapunov exponent and introduce Lyapunov-weighted metrics that provide tighter bounds on predictive disagreement. Leveraging these insights, we develop decision-aligned selection algorithms that choose among near-optimal models based on downstream utility rather than forecast accuracy alone. Extensive experiments on synthetic chaotic systems (Lorenz-96, Kuramoto-Sivashinsky) and real-world applications (wind power, traffic, weather) demonstrate that our framework improves decision quality by 18-34\% while maintaining competitive predictive performance. This work establishes the first rigorous connection between chaos theory and predictive multiplicity, providing principled guidance for deploying machine learning in safety-critical chaotic domains.

arXiv:2605.05217v1 Announce Type: new Abstract: We propose a self-supervised physics-informed neural network (PINN) framework that adaptively balances physics-based and data-driven supervision for scientific machine learning under data scarcity. Unlike prior PINNs that rely on fixed or heuristic weighting of physics residuals and data loss, our approach introduces a learnable blending neuron that dynamically adjusts the relative contribution of each term based on their uncertainties. This mechanism enables stable training and improved generalization without manual tuning. To further enhance efficiency, we integrate a transfer learning strategy that reuses representations from related domains and adapts them to new physical systems with limited data. We validate the framework for the prediction of heat transfer in liquid-metal miniature heat sinks using only 87 CFD datapoints, where the adaptive PINN achieves an error <8%, outperforming shallow neural networks, kernel methods, and physics-only baselines. Our framework provides a general recipe for embedding physics adaptively into neural networks, offering a robust and reproducible approach for data-scarce problems across various scientific domains, including fluid dynamics and material modeling.

arXiv:2605.05213v1 Announce Type: new Abstract: Chronic rhinosinusitis (CRS) is a common heterogeneous inflammatory disorder that causes substantial morbidity and healthcare costs. CRS is difficult to identify early from routine encounters, as symptom presentations overlap with common conditions such as allergic rhinitis, and heterogeneous phenotypes further obscure risk patterns. Prior predictive studies often rely on single-institutional cohorts , which reduce population-level generalizability. To overcome this, we leveraged nationwide longitudinal EHR data from the \textit{All of Us} Research Program to predict CRS diagnosis using two years of pre-diagnostic history. To address extreme feature sparsity and dimensionality in coded EHR data, we implemented a hybrid feature-selection pipeline that combines prevalence-based statistical screening with model-based importance ranking, compressing approximately 110,000 candidate codes into 100 interpretable features. To capture demographic heterogeneity, we trained demographic stratified models across six adult sex and life-stage subgroups with subgroup-specific hyperparameter tuning. Our framework achieved an overall AUC of 0.8461, improving discrimination by 0.0168 over the best baseline. These results demonstrate that routinely collected EHR data may support population-representative CRS risk stratification and inform earlier triage and referral prioritization in primary care.

arXiv:2605.05209v1 Announce Type: new Abstract: Neural networks that land in flat regions of the loss landscape tend to generalise better than those in sharp regions. Sharpness-Aware Minimisation exploits this to improve generalisation. But function-preserving reparameterisation can inflate the Hessian of any minimum by two orders of magnitude without changing a single prediction. If the geometry of weight space can be manufactured from nothing, it cannot be the cause of anything. In other words, flat is simple and simplicity depends on encoding. Here I show that the actual driver is weakness, the volume of completions compatible with the learned function in the learner's embodied language. Weakness is reparameterisation-invariant because it is defined over what the network \emph{does}, not how it is parameterised. I prove weakness is minimax-optimal under exchangeable demands, and that PAC-Bayes bounds work because they correlate with it. On MNIST, the large-batch generalisation advantage \emph{vanishes} as training data grows, from $+1.6\%$ at $n = 2{,}000$ to $+0.02\%$ at $n = 60{,}000$. A quantity whose predictive power depends on how much data you have is not a cause but a confounder. I run head-to-heads on 100 networks with identical architecture and training. For MNIST weakness predicts generalisation ($\rho = +0.374$, $p = 0.00012$), sharpness anticorrelates ($\rho = -0.226$) and simplicity predicts nothing ($p = 0.848$). For Fashion-MNIST ($\rho = +0.384$, $p = 8.15 \times 10^{-5}$), though simplicity is at least somewhat predictive there. Simplicity is dataset dependent, whereas weakness is invariant. Flat minima were never the answer.

arXiv:2605.05219v1 Announce Type: new Abstract: Prefix caching is a key latency optimization for autoregressive LLM serving, yet existing systems assume dense per-token key/value reuse. State-space models change the structure of the problem: a recurrent layer can resume from a single stored state rather than requiring the entire token history. This asymmetry opens a new design point between no reuse and dense caching: store exact recurrent states at a sparse set of checkpoint positions and, on a cache hit, resume from the deepest stored checkpoint and recompute the remaining suffix exactly. We formalize sparse prefix caching as checkpoint placement under a distribution over overlap depths, yielding an exact O(NM) dynamic program. For use cases where requests share a non-trivial prefix (e.g. asking different questions about a single long document), we show that our method consistently improves the Pareto frontier traced by standard heuristics on real-world data. Across QuALITY and System Prompts, distribution-aware placement dominates every fixed-budget baseline on the measured layer-group Pareto frontier and matches or outperforms the strongest heuristic (block caching) while typically using substantially fewer checkpoints, with the largest gains at low checkpoint budgets where the overlap distribution is most non-uniform. The method is most relevant when many requests share a substantial but not identical prefix within a retained cache entry. It preserves exact outputs, does not change the recurrent computation itself or require new recurrent update kernels, applies to recurrent/SSM layers whose hidden state can be extracted and restored exactly, and for hybrid models can be combined with existing KV-cache compression techniques.

arXiv:2605.05220v1 Announce Type: new Abstract: Steering intermediate representations has emerged as a powerful strategy for controlling generative models, particularly in post-deployment alignment and safety settings. However, despite its empirical success, it currently lacks a comprehensive theoretical framework. In this paper, we bridge this gap by formalizing the theory of concept steering. First, we establish a link between steering and affine concept erasure, proving that the standard approach for removing unwanted behaviors is a special case of LEACE (a closed-form method for affine erasure). Next, we formulate a principled theoretical framework for concept switching, LEACE-Switch, and characterize the assumptions under which it provides an optimal affine solution. Building on this analysis, we then introduce MidSteer (Minimal Disturbance concept Steering), a more general affine framework for concept manipulation that relaxes these assumptions and enables directed, minimal-disturbance transformations. We demonstrate that MidSteer performs favorably across a range of tasks, modalities, and architectures, including vision diffusion models and large language models.