Paper Plaza - AcademicHub

01.

arXiv (math.PR) 2026-06-19 DOI: arXiv:2606.19663

Counterexample to a conjecture on the pairwise independent correlation gap using AI

Authors:

arXiv:2606.19663v1 Announce Type: cross Abstract: Aided by the AI tool GPT5.5 Pro, we provide a counterexample to a conjecture made by Ramachandra and Natarajan (2025) [Pairwise independent correlation gap, Operations Research Letters, 107255, 6040].

Read & Discuss → View Source →

02.

arXiv (CS.AI) 2026-06-24 DOI: arXiv:2606.24855

OpenThoughts-Agent: Data Recipes for Agentic Models

Authors:

arXiv:2606.24855v1 Announce Type: new Abstract: Agentic language models dramatically expand the applications of AI yet little is publicly known about how to curate training data for broadly capable agents. Existing open efforts such as SWE-Smith, SERA, and Nemotron-Terminal typically target a single benchmark, leaving open the question of how to train models that generalize across diverse agentic tasks. The OpenThoughts-Agent (OT-Agent) project addresses this gap with a fully open data curation pipeline for training agentic models. We conduct more than 100 controlled ablation experiments to systematically investigate each stage of the pipeline, yielding insights on the importance of task sources and diversity. We then assemble a training set of 100K examples from our pipeline and fine-tune Qwen3-32B on this dataset, which yields an average accuracy of 44.8% across seven agentic benchmarks and a 3.9 percentage point improvement over the strongest existing open data agentic model (Nemotron-Terminal-32B, 40.9%). Moreover, our training data exhibits strong scaling properties, outperforming alternative open datasets at every training set size in compute-controlled comparisons. We publicly release our training sets, data pipeline, experimental data, and models at openthoughts.ai to support future open research on agentic model training.

Read & Discuss → View Source →

03.

arXiv (CS.AI) 2026-06-24 DOI: arXiv:2510.04033

A global log for medical AI

Authors:

arXiv:2510.04033v2 Announce Type: replace Abstract: Modern computer systems rely on syslog, a universal protocol that records critical events across heterogeneous infrastructure. Medicine's rapidly growing AI stack has no equivalent. As medicine deploys AI tools at scale, there is no standard way to record how, when, by whom, and for whom these models are used. Without such records, it is difficult to measure real-world performance and outcomes, detect adverse events, or identify bias and dataset drift. Here we introduce MedLog, a protocol for event-level logging of medical AI. Each time an AI model interacts with a human, another algorithm, or an automated workflow, MedLog creates a record. Each record contains nine core fields: header, model, user, target, inputs, artifacts, outputs, outcomes, and feedback. We apply MedLog across four deployments in the US, Switzerland, and Vietnam: ICU deterioration prediction, tetanus progression monitoring from wearable signals, automated sepsis quality reporting, and patient attendance prediction. MedLog records capture model behavior, workflow interactions, and downstream outcomes, including AI performance degradation during severe weather events in patient attendance prediction and increased laboratory testing after ICU deterioration alerts. MedLog limits the data footprint through risk-based sampling, lifecycle-aware retention policies, and write-behind caching, enabling deployment in low-resource settings. It also supports detailed traces for complex, agentic, or multi-stage workflows, creating a foundation for continuous monitoring, auditing, and improvement of medical AI.

Read & Discuss → View Source →

04.

arXiv (CS.AI) 2026-06-19 DOI: arXiv:2606.19704

Beyond Static Leaderboards: Predictive Validity for the Evaluation of LLM Agents

Authors:

arXiv:2606.19704v1 Announce Type: new Abstract: Agent benchmarks are growing fast, but no single benchmark touches more than four or five of the dimensions that deployment exposes. This paper aggregates the largest coordinated deep-dive of one MCP-based industrial-agent benchmark to date: fourteen parallel implementation studies covering new asset classes (including a multi-modal visual extension), alternative orchestrations, retrieval strategies, reasoning modes, infrastructure optimizations, and evaluation-methodology probes. Consolidating those studies with seven prior agent benchmarks, we argue that aggregate-score leaderboards systematically underspecify deployed-agent evaluation. Rankings derived from aggregate scores do not transfer to out-of-distribution settings; recent public-to-hidden competition retrospectives provide direct empirical evidence of this rank instability. We propose ranking configurations by predictive validity, the correlation between in-sample and out-of-sample rank, rather than in-sample mean, and report a twelve-tier measurement apparatus that exposes the deployment-relevant dimensions HELM and its agent-era successors collapse. The position is operationalized through three falsifiable out-of-distribution criteria with explicit thresholds; existing evidence partly supports it but is too thin to confirm. We close with a pre-registered pilot design and a field-level vision for what the next generation of agentic benchmarks should report.

Read & Discuss → View Source →

05.

medRxiv (Medicine) 2026-06-18 DOI: HASH:ba30e15ac04519cc631eebd7253200ea

Plasma proteomics reveals clinical and mechanistic heterogeneity among individuals who develop coronary artery disease

Authors:

Yang ↗Tan ↗Carrasco-Zanini ↗Su ↗C.-Y ↗Zhou ↗Koyama ↗Natarajan ↗langenberg ↗Lu ↗Yoshiji ↗

BACKGROUND: Individuals who develop coronary artery disease (CAD) are clinically and mechanistically heterogeneous, and understanding this variation is crucial for precise risk stratification and tailored interventions. However, the molecular mechanisms that connect these two kinds of heterogeneity remain unclear, limiting progress toward biologically grounded risk stratification and targeted interventions. Here, we investigated the heterogeneity of individuals who develop CAD by leveraging plasma proteomic signatures, placed individuals along continuous metabolic gradients and revealed the molecular programs underlying these patterns, thereby linking mechanistic variation to clinical heterogeneity. METHODS AND RESULTS: From 42,803 UK Biobank participants, including 3,713 individuals who developed CAD within 10 years (incident CAD), we first identified a 320-protein panel from 2,923 baseline proteins that improved prediction of incident CAD beyond clinical risk scores. Using reverse graph embedding, we reduced the proteomic data to two dimensions and mapped each incident case onto the resulting two-dimensional latent proteomic space. These proteomic dimensions show significant associations with cardiometabolic and kidney-related clinical markers. The patterns were replicated in the EPIC-Norfolk study. Phenome-wide Cox regression analyses further linked these proteomic dimensions to 10-year incidence rates for various diseases, including type 2 diabetes, obesity, and chronic kidney disease (CKD). Furthermore, adding the proteomic dimensions to clinical variable-based Cox regression model improved prediction of 10-year incidence of CKD and other diseases, demonstrating the value of proteomic dimensions beyond conventional clinical risk factors. Moreover, individuals with prevalent CAD (diagnosed before proteomic sampling) exhibited high, metabolically adverse dimension values, indicating that these axes capture cumulative metabolic burden. Pathway enrichment analyses implicated altered extracellular matrix organization and immune programs among the proteins contributing to the proteomic dimensions. CONCLUSIONS: Our findings demonstrate that plasma proteomic signatures can dissect the heterogeneity of individuals who develop CAD in continuous phenotypic gradients, improve prediction of CAD and comorbidities, and map underlying biological mechanisms.

Read & Discuss → View Source →

06.

arXiv (CS.AI) 2026-06-19 DOI: arXiv:2602.01425

One Probe Won't Catch Them All: Towards Targeted Deception Detection

Authors:

Vikram Natarajan ↗Devina Jain ↗Shivam Arora ↗Satvik Golechha ↗Joseph Bloom ↗

arXiv:2602.01425v2 Announce Type: replace Abstract: Linear probes are a promising approach for monitoring AI systems for deceptive behaviour. Previous work has shown that a linear classifier trained on a contrastive instruction pair and a simple dataset can achieve good performance. However, these probes exhibit notable failures even in straightforward scenarios, including spurious correlations and false positives on non-deceptive responses. In this paper, we demonstrate that deception detection is inherently heterogeneous: while a single universal probe achieves modest improvements (+0.032 AUC), post-hoc oracle analysis reveals substantially higher potential (+0.108 AUC) when probes are matched to specific deception types, and synthetic validation experiments suggest this ceiling is achievable a priori when the deception type is known in advance. Our findings reveal that instruction pairs capture deceptive intent rather than content-specific patterns, explaining why prompt choice dominates probe performance (70.6% of variance). Given this heterogeneity, we conclude that organizations should define their specific threat models and deploy appropriately matched probes rather than seeking a universal deception detector.

Read & Discuss → View Source →

07.

arXiv (CS.CL) 2026-06-11 DOI: arXiv:2606.11429

Gumbel-BEARD: Automatic Layer Selection for Self-Supervised Adaptation of Whisper in Low-Resource Domains

Authors:

Zilai Wang ↗Natarajan Balaji Shankar ↗Mohan Shi ↗Kaiyuan Zhang ↗Abeer Alwan ↗

Speech foundation models often struggle in low-resource domains due to domain mismatch and data scarcity. We propose Gumbel-BEARD, a domain adaptation framework that automates Whisper encoder layer selection via an end-to-end trainable hard Gumbel-Softmax selector. It enables self-supervised adaptation with a BEST-RQ objective that dynamically adapts to target acoustic characteristics without manual tuning. Experiments on the MyST child speech corpus demonstrate efficiency and scalability: with 10 h of labeled data for fine-tuning, our method matches a fully supervised baseline trained on the complete 133 h labeled set. We establish new state-of-the-art word error rates (WERs) of 8.21% using Whisper-medium on MyST and 11.06% using Whisper-small on the OGI Spontaneous dataset. Evaluation on CORAAL further confirms robustness to adult dialectal domain shifts, with up to 6% relative WER reduction, highlighting the generalizability of our approach to diverse low-resource conditions.

Read & Discuss → View Source →

08.

arXiv (quant-ph) 2026-06-25 DOI: arXiv:2605.26096

Rounding Almost Commuting Hamiltonians

Authors:

Islam Faisal ↗Anand Natarajan ↗Alexander Poremba ↗

arXiv:2605.26096v2 Announce Type: replace Abstract: Commuting Hamiltonians lie at the boundary between classical constraint satisfaction and quantum many-body physics, exhibiting rich quantum structure while remaining more tractable than general noncommuting models. In contrast, physical Hamiltonians are rarely exactly commuting, which naturally motivates the study of almost commuting Hamiltonians. Despite their relevance, the implications of approximate commutation are only poorly understood. In this work, we show how to efficiently approximate any almost commuting $2$-local qubit Hamiltonian by a commuting one: we give a new locality-preserving algorithmic rounding technique that maps any $2$-local Hamiltonian $H=\sum_{i=1}^m h_i$ with $\|[h_i,h_j]\| \leq \epsilon$ to a nearby Hamiltonian $\hat{H}$ whose terms pair-wise commute, and which is within overall distance $\|H-\hat{H}\| = O(m\,\epsilon^{1/6})$. As a consequence, we show that $\delta$-approximations to the ground energy for $\epsilon$-almost commuting $2$-local qubit Hamiltonians lie in $\mathsf{NP}$ when $\delta \gg m\epsilon^{1/6}$, extending the classical containment well beyond the commuting setting. Finally, we present two applications of our rounding framework: Gibbs sampling and fast Hamiltonian simulation for almost commuting systems.

Read & Discuss → View Source →

09.

medRxiv (Medicine) 2026-06-11 DOI: HASH:913ec6424b476b1b5739bce1b64a1534

Long-term Penetrance of Disease Variants in Genes Prioritized for Genomic Newborn Screening: Evidence from Adult Biobanks

Authors:

Gold ↗N. B ↗Zouk ↗Yeo ↗Lipsitz ↗Koyama ↗Somanchi ↗Perez ↗Selvaraj ↗M. S ↗O'Grady ↗Miller ↗…

Importance: Genomic newborn screening (gNBS) is a potential public health intervention, but its positive predictive value (PPV) remains uncertain. Estimating the prevalence and penetrance of pathogenic and likely pathogenic (P/LP) variants in genes prioritized for screening may clarify the long-term PPV and clinical utility of gNBS. Objective: To compare ICD-based ascertainment, electronic medical record (EMR) review, and clinical assessment of genetic disorders in adults with P/LP variants in 54 genes prioritized for gNBS. Design: Two-cohort observational study with EMR review and clinical assessment in the hospital-based cohort. Setting: The U.K. Biobank (UKB) and Mass General Brigham Biobank (MGBB). Participants: 451,877 adults from the UKB and 53,371 from the MGBB, all with exome sequencing data. Exposures: P/LP variants in 54 genes prioritized through expert consensus for gNBS, in genotypes consistent with each gene's inheritance pattern. Main outcomes and measures: The primary outcome was the absolute difference in the proportion of MGBB participants identified as affected by ICD versus EMR ascertainment. Secondary outcomes included findings from clinical assessments of undiagnosed MGBB participants, corrected UKB penetrance estimates, and extrapolation to U.S.. annual birth cohorts and living adults. Results: P/LP variants were identified in 665 UKB participants (0.15%) and 82 MGBB participants (0.15%), approximately 1 in 650. In MGBB, EMR review revealed that 58/82 individuals (70.7%) were undiagnosed, although 25 of 58 (43.1%) had documented symptoms. Disease-associated ICD codes were found in 39.0% (32/82) of participants, whereas EMR review identified symptoms in 59.8% (49/82, McNemar P

Read & Discuss → View Source →

10.

medRxiv (Medicine) 2026-06-25 DOI: HASH:32130886ed3a19694e5aa2bded5af853

Performance of Cardiovascular Polygenic Risk Scores in Carotid Stenosis Identification

Authors:

Bellomo ↗T. R ↗Misra ↗Cai ↗Sanchez Garcia ↗Patel ↗A. P ↗Flores ↗A. M. T ↗Nordan ↗Nakao ↗Yu ↗…

Background: Clinically significant carotid stenosis remains a major cause of ischemic stroke (IS), yet prediction of disease progression is limited. Polygenic risk scores (PRSs) for coronary artery disease (CAD) and peripheral artery disease (PAD) have demonstrated associations with atherosclerosis burden and major cardiovascular disease (CVD) events, but whether these insights extend to carotid stenosis is unclear. We evaluated the association and discriminative performance of validated PRSs for CAD, PAD, IS, and carotid intima-media thickness (cIMT) with carotid stenosis. Methods: Carotid stenosis was identified in genotyped Mass General Brigham Biobank participants using validated ICD- and CPT-based phenotyping algorithms. Logistic regression adjusted for age, sex, and 10 ancestry principal components assessed PRS associations. Incremental discrimination was evaluated using changes in Harrell's C-statistic. Results: Compared with 52,636 controls, 670 participants with carotid stenosis were more frequently male (61.5% vs 44.1%), older (70.8 SD 9.0 vs 53.4 SD 17.2 years), and more likely to be European (95.7% vs 84.1%). The IS (OR 1.31, 95% CI 1.21?1.41), CAD (OR 1.62, 95% CI 1.50?1.75), and PAD (OR 1.66, 95% CI 1.54?1.80) PRSs were each associated with carotid stenosis (all p

Read & Discuss → View Source →

Explore the Frontier of Global Academia