Academic Intelligence · Curated Daily

Explore the Frontier of Global Academia

AcademicHub aggregates real-time literature from top journals and preprint platforms. Build your personal research radar and let large language models compile cross-disciplinary analysis briefings automatically.

01.
medRxiv (Medicine) 2026-06-22

The circulating blood proteome of childhood acute leukemia

The circulating blood proteome provides a systemic readout of disease biology and holds promise for advancing diagnostics and disease monitoring in pediatric leukemia. Here, we profiled 3072 proteins in diagnostic serum from 54 children with acute lymphoblastic leukemia (ALL), 21 with acute myeloid leukemia (AML), and 12 healthy controls using the Olink Proximity Extension Assay. We observed profound alterations in circulating protein levels in leukemia patients compared with controls and identified immunophenotype-specific proteins, including SIGLEC15 in B-cell precursor ALL (BCP-ALL), NOTCH1 in T-ALL, and CEBPA in AML, all which remained high even in patients with low (

02.
arXiv (CS.LG) 2026-06-11

Re-evaluating Confidence Remasking in Masked Diffusion Language Models

arXiv:2606.12232v1 Announce Type: new Abstract: Masked diffusion language models (dLLMs) have recently emerged as a competitive alternative to autoregressive language models, with the promise of faster inference via parallel token generation. A notable limitation of the masked formulation, however, is that once a token has been unmasked it can no longer be revised, leaving dLLMs vulnerable to early sampling mistakes. To address this, a growing body of work has sought to extend masked dLLMs with self-correcting (remasking) capabilities. One appealing subset of these methods does so in a training-free, post-hoc manner based on token confidences, with encouraging early reported results. In this work, we revisit the empirical evaluation of a representative post-hoc remasking method, WINO [Hong et al., 2026], and find that under standard decoding settings (shorter block lengths) it brings little-to-no benefit over confidence-based unmasking alone [Wu et al., 2025]. Extending the evaluation to non-greedy decoding, we find that while confidence-based remasking can mitigate errors introduced by increased stochasticity to some extent, it also exacerbates the diversity collapse previously reported for confidence-based unmasking. Overall, our results show that the benefits of post-hoc confidence-based remasking are highly setting-dependent, underscoring the need for a more comprehensive evaluation framework.

03.
arXiv (CS.CV) 2026-06-15

A Lightweight Fiducial-Based Pipeline for 3D Hyperspectral Mapping of ex-vivo Lumpectomy Specimens

Hyperspectral Imaging (HSI) is a promising modality for intraoperative assessment of resection margins in Breast-Conserving Surgery (BCS), but its clinical translation requires aligning the inherently 2D spectral information onto the 3D shape of the excised tissue so that suspicious regions can be precisely localized for targeted follow-up. We present a fully automated, calibration-free pipeline that produces a 3D hyperspectral point cloud of an ex-vivo lumpectomy specimen from a set of consumer-camera RGB images and a single top-down HSI acquisition. The 3D geometry is reconstructed with a deep-learning Structure-from-Motion backbone, stabilized in a metric reference frame by a custom bundle adjustment that enforces consistency on the corners of four ArUco markers placed around the specimen. The HSI cube is then registered to the reconstruction without recovering the HSI camera pose: the markers, visible in both modalities, define 16 corner correspondences that drive a planar homography, and 3D coordinates are recovered by lookup on an orthographically rendered depth map. Evaluated on two ex-vivo lumpectomy specimens, the pipeline achieves a median 3D registration error below 1~mm and a 2D reprojection error below 0.02 mm, with a total per-specimen processing time under 4 minutes on accelerated hardware. These results support the feasibility of integrating HSI-guided spatial localization into intraoperative margin assessment workflows for breast-conserving surgery.

04.
arXiv (CS.CV) 2026-06-11

Bridging the Modality Gap in Forensic Image Retrieval

Automated image retrieval plays an increasingly critical role in modern forensic analysis, supporting investigative workflows that rely on efficient comparison of visual evidence. While prior work has focused primarily on developing and optimizing multimodal retrieval systems, limited attention has been paid to evaluating the forensic applicability of these technologies across diverse real-world scenarios. In this study, we present a unified retrieval framework adapted to four key forensic tasks: (1) tattoo image retrieval given a tattoo query image; (2) tattoo retrieval guided by human-expert textual descriptions, modelling the common situation where a witness verbally describes a tattoo; (3) tattoo retrieval from hand-drawn sketches; and (4) face retrieval from forensic face sketches. Our system leverages a multimodal large language model (MLLM) to automatically generate structured textual descriptions for all queries and gallery images, followed by sentence-transformer embedding for text-based comparison. We evaluate retrieval using visual-only embeddings, text-only embeddings and a multimodal fusion strategy that combines text- and image-based similarity scores derived from state-of-the-art visual feature extractors relevant to each task. The fusion of modalities consistently improves retrieval precision and robustness, especially in scenarios where visual information is limited or noisy (e.g., sketches, partial tattoos, or fragmented witness statements). This work highlights the forensic value of a unified multimodal retrieval pipeline and demonstrates how modern MLLMs can operationalize challenging forensic tasks that traditionally rely on manual expert analysis. Our results position multimodal retrieval as a promising tool for supporting investigative workflows involving tattoos, facial composites, and witness descriptions.

05.
arXiv (CS.LG) 2026-06-19

EQPO: Equitable Group Relative Policy Optimization for Clinical Reasoning

arXiv:2510.19893v2 Announce Type: replace Abstract: Medical AI systems demonstrated impressive diagnostic performance, yet they routinely show uneven accuracy across demographic groups, disadvantaging underrepresented populations. Although multimodal reasoning foundation models have pushed clinical diagnosis forward, reinforcement learning-based post-training tends to absorb and magnify the biases present in majority-dominated training corpora. We propose Equitable Group Relative Policy Optimization (EQPO), a hierarchical reinforcement learning method that encourages balanced learning across heterogeneous clinical populations by adaptively reweighting samples according to subgroup representation, task difficulty, and data source. As demographic annotations are frequently missing in real-world clinical data, EQPO additionally applies unsupervised clustering to recover latent subpopulations when they are unavailable. On 7 diagnostic benchmarks covering 5 modalities (X-ray, CT, dermoscopy, mammography, ultrasound), EQPO reduces F1 standard deviation by 43.9% and the maximum cross-group F1 gap by 42.7% on QoQ-Med3-8B over vanilla GRPO, and narrows predictive parity gaps by 27.2% on MedGemma-4B over bias-mitigated RL baselines while raising F1 by 12.5% even without any demographic labels. Examining the training trajectory shows that EQPO steadily improves fairness over the course of optimization, in contrast to baseline methods whose fairness degrades as training proceeds, and the discovered implicit groups remain stable and align with masked demographic attributes. We further release EquiMedGemma-4B and EquiQoQ-Med3-8B, equitability-aware clinical VLLMs that attain state-of-the-art accuracy with markedly smaller demographic gaps.

06.
arXiv (math.PR) 2026-06-16

Malliavin Calculus for the stochastic Cahn-Hilliard equation driven by fractional noise

arXiv:2601.10490v2 Announce Type: replace Abstract: The stochastic partial differential equation analyzed in this work is the Cahn-Hilliard equation perturbed by an additive fractional white noise (fractional in time and white in space). We work in the case of one spatial dimension and apply Malliavin calculus to investigate the existence of a density for the stochastic solution $u$. In particular, we show that $u$ admits continuous paths almost surely and construct a localizing sequence through which we prove that its Malliavin derivative exists locally, and that its law is absolutely continuous with respect to the Lebesgue measure on $\bf R$, establishing thus that a density exists. A key contribution of this work is the analysis of the stochastic integral appearing in the mild formulation: we derive sharp estimates for the expectation of the $p$-th power ($p \geq 2$) of the $L^{\infty}(D)$-norm of this stochastic integral as well as for the integral involving the $L^{\infty}(D)$-norm of the operator associated with the kernel appearing in the integral representation of the fractional noise, all of which are essential for this study.

07.
arXiv (CS.LG) 2026-06-11

Coverage Guarantees for Pseudo-Calibrated Conformal Prediction under Distribution Shift

arXiv:2602.14913v2 Announce Type: replace Abstract: Conformal prediction (CP) offers distribution-free marginal coverage guarantees under an exchangeability assumption, but these guarantees can fail if the data distribution shifts. We analyze the use of pseudo-calibration as a tool to counter this performance loss under a bounded label-conditional covariate shift model. Using tools from domain adaptation, we derive a lower bound on target coverage in terms of the source-domain loss of the classifier and a Wasserstein measure of the shift. Using this result, we provide a method to design pseudo-calibrated sets that inflate the conformal threshold by a slack parameter to keep target coverage above a prescribed level. Finally, we propose a source-tuned pseudo-calibration algorithm that interpolates between hard pseudo-labels and randomized labels as a function of classifier uncertainty. Numerical experiments show that our bounds qualitatively track pseudo-calibration behavior and that the source-tuned scheme mitigates coverage degradation under distribution shift while maintaining nontrivial prediction set sizes.

08.
arXiv (CS.AI) 2026-06-24

Neural Network-Based Parametric Model Reduction for Predicting Turbulent Flow for Different Vehicle Geometries

arXiv:2606.24265v1 Announce Type: cross Abstract: Numerical simulations in industrial applications often require performing numerous high-precision computations parameterized by specific experimental conditions. For instance, in vehicle body design, aerodynamic simulations are essential for evaluating the aerodynamic characteristics of various proposed body geometries. However, computational resource constraints often become a bottleneck. Therefore, achieving the desired accuracy while minimizing computational cost is crucial. To address this challenge, model reduction methods have been developed to decrease the degrees of freedom by constraining the possible states of a physical system to a lower-dimensional subspace. In particular, reduction techniques that project the system onto a nonlinear subspace using neural networks have been actively studied. Our previous research developed a reduced-order model that integrates neural-network-based model reduction with a time-evolution method, implemented as a distributed parallel training framework to process high-resolution flow field data efficiently. In this study, we extend this reduction approach by incorporating a variational autoencoder to assess its robustness in high-Reynolds-number flows around multiple vehicle bodies with varying geometries. Specifically, we evaluate the reconstruction accuracy of vortex generation across different spatial and temporal scales using a compact latent representation, with a particular focus on the flow behavior near the rear end of the vehicle body.

09.
medRxiv (Medicine) 2026-06-15

Neural Correlates of Human Food Memory link to Microbial, Homeostatic, and Hedonic Signals: Evidence from a Prebiotic Randomized Clinical Trial

Background Homeostatic and hedonic brain circuits regulate eating behavior but also shape how food memories are encoded and retrieved. Objective We examined neural correlates during food memory encoding and retrieval during functional MRI before and after a 14-day prebiotic intervention in a preregistered, double-blind crossover trial (NCT03829189). Design 55 healthy adults with overweight (19 females, age 28{+/-}6.5, BMI 25-30 kg/m2) underwent 3 Tesla task-based functional MRI before and after dietary intervention of prebiotic (30g inulin/day) or equicaloric placebo for 14 days. Peripheral metabolic, short-chain fatty acids (SCFA), and microbial markers using 16S rRNA analysis were assessed in fasting blood and feces. Results Food memory was enhanced by assigned reward value and engaged brain activity in hedonic regions, including the nucleus accumbens, orbitofrontal cortex, caudate, cingulate, dorsomedial prefrontal cortex, and ventral tegmental area, as well as homeostatic and memory-related such as the hypothalamus and the hippocampus. Higher neural activations during food encoding were related to higher Actinobacteriota abundance, fecal SCFA acetate, and creatinine levels, and lower ghrelin levels. Activations in reward-related and homeostatic brain areas partially correlated with insulin, glucagon-like peptide-1, leptin, and thyroid-stimulating hormone levels. Neural activations related to food memory decreased after prebiotic intervention. The prebiotic supplementation induced decrease of hippocampal activity during food encoding related to changes in gut microbiota Firmicutes abundance. Conclusions This study indicates that neuronal food-related memory processes depend on homeostatic and hedonic brain signals modulated by the gut-brain axis. Our findings raise implications for the treatment of obesity and substance use disorder.

10.
bioRxiv (Bioinfo) 2026-06-16

Super Learner Ensemble Modeling of CPTAC Proteomic Data for Survival Prediction in Head and Neck Squamous Cell Carcinoma

Survival analysis in head and neck squamous cell carcinoma (HNSCC) is traditionally performed using Cox proportional hazards models, alongside some exploration into black-box machine learning methods. The Super Learner (SL) algorithm addresses this model selection dilemma by combining diverse candidate algorithms into a weighted ensemble to perform comparably to the best candidate method. This study evaluates the performance of SL in HNSCC. Proteomic features as well as clinical covariates from 96 CPTAC HNSCC samples were modeled with three candidate algorithms (Cox LASSO, Cox Ridge, and Random Survival Forest) as well as the ensemble SL method. Models were optimized via Uno's time-dependent Concordance Index (C-index) and tested at 1- and 3-year time horizons using 2000 bootstrap resamples. The Cox Ridge regression model achieved the highest predictive accuracy among the four total methods. However, the SL demonstrated stable performance over both time horizons (1-year C-index: 0.985; 3-year C-index: 0.960). Variable importance analysis of the Cox Ridge model successfully identified malignant proteins (ATR, MAML1, MIEN1) alongside novel potential prognostic indicators (ZNF800, KERA). This analysis emphasizes the statistical necessity for larger cohorts for ensemble learning, while providing a benchmark of proteomic indicators in HNSCC.

11.
medRxiv (Medicine) 2026-06-11

Computer Vision Scoring of Figure Copy and Recall

Objective. Figure copy and recall tests are sensitive measures of visuoconstruction and visual episodic memory, but their clinical is constrained by labor-intensive manual scoring. We developed and validated an automated, element-level scoring pipeline using Vertex AI object detection for the tablet-based figure copy and recall tasks in the California Cognitive Assessment Battery (CCAB). The automated scoring pipeline duplicated the scoring procedures used by expert manual raters. Methods. A normative sample of 2,011 community-dwelling adults aged 18-90 completed figure copy and delayed recall trials at baseline, with subsamples retested at 1 day and at 6, 18, and 30 months. Participants completed the drawings with their index finger on a tablet computer with finger position digitized to analyze the speed and timing of individual drawing strokes A convolutional object-detection model trained on the Vertex AI AutoML Vision platform identified each of twelve canonical figure elements in rendered drawings. Separate element presence and location scores were computed after homographically warping drawings onto a canonical template to produce trial-level Element, Location, and Total scores. To compare Vertex and human scores, Vertex AI and expert human raters independently scored 1500 randomly selected drawings to evaluate inter-rater agreement, including a common subset of 100 drawings scored by Vertex AI and all raters. Results. Total scores were virtually indistinguishable (r = 0.966) from human-human agreement (mean r = 0.971) as were Element presence scores (mean r = 0.959 vs. r = 0.963). Location-score agreement (r = 0.951) was slightly below the human-human mean (r = 0.972) due to pixel-level analysis by Vertex AI that was impossible for human raters. The Vertex pipeline showed no preferential advantage for the single expert rater who categorized Elements during training. Automated scores showed strong demographic gradients, age effects on Recall (r = -0.32) were approximately twice those in Copy conditions (r = -0.16). A Memory Cost score (Recall - Copy) showed a monotonic age-related decline from +0.40 z in the youngest subjects to -0.54 z in the oldest. Kinetic analysis revealed that drawing speed and efficiency showed significant age-related changes. Overnight test-retest reliability was high (Recall r = 0.72) and the Recall trial showed a large overnight learning effect ({Delta} = +1.18) that continued with repeated tests up to 30 months ({Delta} = +0.75).

12.
arXiv (CS.AI) 2026-06-19

The Autonomy Tax: Defense Training Breaks LLM Agents

arXiv:2603.19423v2 Announce Type: replace-cross Abstract: Large language model (LLM) agents increasingly rely on external tools (file operations, API calls, database transactions) to autonomously complete complex multi-step tasks. Practitioners deploy defense-trained models to protect against prompt injection attacks that manipulate agent behavior through malicious observations or retrieved content. We reveal a fundamental capability-alignment paradox: defense training designed to improve safety systematically destroys agent competence while failing to prevent sophisticated attacks. Evaluating defended models against undefended baselines across 97 agent tasks and 1,000 adversarial prompts, we uncover three systematic biases unique to multi-step agents. Agent incompetence bias manifests as immediate tool execution breakdown, with models refusing or generating invalid actions on benign tasks before observing any external content. Cascade amplification bias causes early failures to propagate through retry loops, pushing defended models to timeout on 99\% of tasks compared to 13\% for baselines. Trigger bias leads to paradoxical security degradation where defended models perform worse than undefended baselines while straightforward attacks bypass defenses at high rates. Root cause analysis reveals these biases stem from shortcut learning: models overfit to surface attack patterns rather than semantic threat understanding, evidenced by extreme variance in defense effectiveness across attack categories. Our findings demonstrate that current defense paradigms optimize for single-turn refusal benchmarks while rendering multi-step agents fundamentally unreliable, necessitating new approaches that preserve tool execution competence under adversarial conditions.

13.
arXiv (CS.CL) 2026-06-17

Top-Theta Attention: Sparsifying Transformers by Compensated Thresholding

We present Top-Theta (Top-$\theta$) Attention, a training-free method for sparsifying transformer attention during inference. Our key insight is that static, per-head thresholds can be calibrated to retain the desired constant number of significant elements per attention row. This approach enables content-based sparsity without retraining, and it remains robust across data domains. We further introduce compensation techniques to preserve accuracy under aggressive sparsification, establishing attention thresholding as a practical and principled alternative to top-k attention. We provide extensive evaluation on natural language processing tasks, showing that Top-$\theta$ achieves 3-10x reduction in V-cache usage and up to 10x fewer attention elements during inference while degrading no more than 1% in accuracy.

14.
arXiv (CS.AI) 2026-06-17

Descriptor: Certus Caliber Classification Gunshot Dataset (C3GD)

arXiv:2606.18135v1 Announce Type: cross Abstract: In this work, we introduce the Certus Caliber Classification Gunshot Dataset (C3GD), a publicly accessible data set developed for the analysis of firearm muzzle blast sounds. The dataset aims to provide a wide variety of firearms, calibers, cartridges, microphones, and microphone locations with metadata detailed beyond what is currently otherwise available. It comprises more than 8000 field-collected data points from 28 firearms across 16 calibers. Because data collection in the field is costly, much of the existing research has been done using gunshot audio collected from the internet, which increases the risk of low-quality data and label noise. This dataset is primarily focused on caliber classification, but can also be used for gunshot detection, audio separation, and audio signal processing, providing a diversified and real-world reference. The dataset aims to provide enough diversity to be able to generalize to more real-world applications while also providing enough metadata for detailed academic analysis.

15.
arXiv (CS.LG) 2026-06-24

ParallelBench: Understanding the Trade-offs of Parallel Decoding in Diffusion LLMs

arXiv:2510.04767v2 Announce Type: replace Abstract: While most autoregressive LLMs are constrained to one-by-one decoding, diffusion LLMs (dLLMs) have attracted growing interest for their potential to dramatically accelerate inference through parallel decoding. Despite this promise, the conditional independence assumption in dLLMs causes parallel decoding to ignore token dependencies, inevitably degrading generation quality when these dependencies are strong. However, existing works largely overlook these inherent challenges, and evaluations on standard benchmarks (e.g., math and coding) are not sufficient to capture the quality degradation caused by parallel decoding. To address this gap, we first provide an information-theoretic analysis of parallel decoding. We then conduct case studies on analytically tractable synthetic list operations from both data distribution and decoding strategy perspectives, offering quantitative insights that highlight the fundamental limitations of parallel decoding. Building on these insights, we propose ParallelBench, the first benchmark specifically designed for dLLMs, featuring realistic tasks that are trivial for humans and autoregressive LLMs yet exceptionally challenging for dLLMs under parallel decoding. Using ParallelBench, we systematically analyze both dLLMs and autoregressive LLMs, revealing that: (i) dLLMs under parallel decoding can suffer dramatic quality degradation in real-world scenarios, and (ii) current parallel decoding strategies struggle to adapt their degree of parallelism based on task difficulty, thus failing to achieve meaningful speedup without compromising quality. Our findings underscore the pressing need for innovative decoding methods that can overcome the current speed-quality trade-off. We release our benchmark to help accelerate the development of truly efficient dLLMs.

16.
arXiv (CS.AI) 2026-06-16

LabOSBench: Benchmarking Computer Use Agents for Scientific Instrument Control

arXiv:2606.16802v1 Announce Type: new Abstract: Current computer-use benchmarks primarily focus on software operation tasks in virtualized systems, whereas scientific instrumentation scenarios require coordinated control over complex interfaces, and feedback-driven parameter adjustment. However, directly evaluating agents on physical high-precision instruments is impractical due to high cost, safety risks, limited accessibility, and difficulty in ensuring reproducible evaluation. This motivates the need for a simulated yet realistic testbed that preserves the operational challenges of scientific instruments while enabling scalable and safe benchmarking. To this end, we introduce LabOSBench, a challenging benchmark for multimodal GUI agents built on a suite of web-based scientific-instrument simulators. Operating directly via a browser, LabOSBench avoids resource-heavy OS virtualization while supporting flexible task configuration and execution-based evaluation. Specifically, LabOSBench constructs 96 subtasks across eight instrument simulators, covering workflows from sample loading, alignment, parameter tuning, and data acquisition to result inspection. We evaluate general-purpose vision-language models, specialized GUI agent models, and advanced agentic frameworks at both subtask and end-to-end levels. Our experiments reveal that while existing agents can complete many structured GUI subtasks, they still struggle with feedback-driven operations and long-horizon workflow execution. Overall, LabOSBench provides a reproducible, low-cost testbed for advancing computer-using agents toward scientific-instrument control.

17.
arXiv (CS.CL) 2026-06-11

Hubs or Fringes: Pretraining Data Selection via Web Graph Centrality

The performance of modern language models depends critically on pretraining data composition. Yet existing data selection methods rely on auxiliary classifiers for document scoring or mixture optimization, adding computational overhead and dependence on labeled data. We propose WebGraphMix, a lightweight data selection framework that computes structural centrality scores over the Common Crawl host-level web graph and uses them to vary the proportion of central versus peripheral documents in the pretraining mixture. We hypothesize that central hosts expose models to reusable abstractions, while peripheral hosts encode specialized, long-tail knowledge. WebGraphMix computes centrality scores efficiently at web scale, requiring no model training, labeled data, or downstream supervision. We integrate WebGraphMix into the DataComp-LM pipeline and train models at 400M and 1B parameter scales with 8B and 28B tokens respectively, evaluating on 23 tasks ranging from factual knowledge to symbolic reasoning. Our experiments show that central and peripheral web regions encode complementary capabilities. Mixture combining both at a ratio of 1:1 achieves 41.4% on average, compared to 39.8% for uniform sampling. Combining structural scores with document-level quality classifier scores further improves performance to 43.8%. These findings demonstrate that web graph topology is a meaningful axis for pretraining data curation, capturing information that is largely orthogonal to existing content-based approaches.

18.
arXiv (CS.LG) 2026-06-19

A Unified Perspective on the Dynamics of Deep Transformers

arXiv:2501.18322v2 Announce Type: replace Abstract: Transformers, which are state-of-the-art in most machine learning tasks, represent the data as sequences of vectors called tokens. This representation is then exploited by the attention function, which learns dependencies between tokens and is key to the success of Transformers. However, the iterative application of attention across layers induces complex dynamics that remain to be fully understood. To analyze these dynamics, we identify each input sequence with a probability measure and model its evolution as a Vlasov equation called Transformer PDE, whose velocity field is non-linear in the probability measure. Our first set of contributions focuses on compactly supported initial data. We show the Transformer PDE is well-posed and is the mean-field limit of an interacting particle system, thus generalizing and extending previous analysis to several variants of self-attention: multi-head attention, L2 attention, Sinkhorn attention, Sigmoid attention, and masked attention–leveraging a conditional Wasserstein framework. In a second set of contributions, we are the first to study non-compactly supported initial conditions, by focusing on Gaussian initial data. Again for different types of attention, we show that the Transformer PDE preserves the space of Gaussian measures, which allows us to analyze the Gaussian case theoretically and numerically to identify typical behaviors. This Gaussian analysis captures the evolution of data anisotropy through a deep Transformer. In particular, we highlight a clustering phenomenon that parallels previous results in the non-normalized discrete case.

19.
arXiv (CS.AI) 2026-06-16

JADE: Expert-Grounded Dynamic Evaluation for Open-Ended Professional Tasks

arXiv:2602.06486v2 Announce Type: replace Abstract: Evaluating agentic AI on open-ended professional tasks faces a fundamental dilemma between rigor and flexibility. Static rubrics provide rigorous, reproducible assessment but fail to accommodate diverse valid response strategies, while LLM-as-a-judge approaches adapt to individual responses yet suffer from instability and bias. Human experts address this dilemma by combining domain-grounded principles with dynamic, claim-level assessment. Inspired by this process, we propose JADE, a two-layer evaluation framework. Layer 1 encodes expert knowledge as a predefined set of evaluation skills, providing stable evaluation criteria. Layer 2 performs report-specific, claim-level evaluation to flexibly assess diverse reasoning strategies, with evidence-dependency gating to invalidate conclusions built on refuted claims. Experiments on BizBench show that JADE improves evaluation stability and reveals critical agent failure modes missed by holistic LLM-based evaluators. We further demonstrate strong alignment with expert-authored rubrics and effective transfer to HealthBench and DR.BENCH, covering medical and 10-domain professional evaluation settings. Code and data are available at https://github.com/smiling-world/JADE.

20.
arXiv (CS.CV) 2026-06-12

Amnesia: A Stealthy Replay Attack on Continual Learning Dreams

Continual learning (CL) models often use experience replay to reduce catastrophic forgetting, but their robustness to replay sampling interference remains underexplored. Existing CL attacks alter inputs or training pipelines (poisoning/backdoors) and rarely include explicit auditable constraints, limiting realism. Here, auditability means a monitor can verify compliance from sampler-visible telemetry - e.g., logged replay index/label statistics - by checking that the realized replay class histogram stays close to a nominal baseline and that replay rate is unchanged per batch and/or over a rolling window. We study a limited-privilege insider who controls only replay index selection, not pixels, labels, or model parameters, while staying within auditable limits such as queue priorities. We introduce Amnesia, a replay composition attack that maximizes degradation under two budgets: a visibility budget delta bounding the TV/KL divergence from a nominal class histogram p0, and a mass budget f fixing the replay rate. Amnesia has two steps: (i) compute lightweight class utilities, such as EMA loss or confidence, to tilt p0 toward harmful classes; and (ii) project the tilt back into the delta-ball using efficient KL (exponential tilt) or TV (balanced mass redistribution) optimizers. A windowed scheduler enforces rolling audits. Across challenging CL benchmarks and strong replay baselines, Amnesia consistently lowers final accuracy (ACC) and worsens backward transfer (-BWT). The KL variant delivers high impact while remaining largely undetected under multiple audit schemes, including per-batch and rolling-window checks. The TV variant is more damaging but easier to detect, especially under tight per-class constraints. These results expose index-only replay control as a practical, auditable threat surface in CL systems and establish a principled impact-visibility trade-off.

21.
arXiv (CS.CL) 2026-06-15

Did You Forget What I Asked? Prospective Memory Failures in Large Language Models

Authors:

Large language models often fail to satisfy formatting instructions when they must simultaneously perform demanding tasks. We study this behaviour through a prospective memory inspired lens from cognitive psychology, using a controlled paradigm that combines verifiable formatting constraints with benchmark tasks of increasing complexity. Across three model families and over 8,000 prompts, compliance drops by 2-21% under concurrent task load. Vulnerability is highly type-dependent: terminal constraints (requiring action at the response boundary) degrade most, with drops up to 50%, while avoidance constraints remain comparatively robust. A salience-enhanced format (explicit instruction framing plus a trailing reminder) recovers much of the lost compliance, restoring performance to 90-100% in many settings. Interference is bidirectional: formatting constraints can also reduce task accuracy, with one model's GSM8K accuracy dropping from 93% to 27%. In additional stacking experiments, joint compliance declines sharply as constraints accumulate. All results use deterministic programmatic checkers without an LLM-as-judge component on publicly available datasets.

22.
arXiv (CS.AI) 2026-06-12

The Query Channel: Information-Theoretic Limits of Masking-Based Explanations

arXiv:2604.16689v2 Announce Type: replace Abstract: Masking-based post-hoc explanation methods, such as KernelSHAP and LIME, estimate local feature importance by querying a black-box model under randomized perturbations. This paper formulates this procedure as communication over a query channel, where the latent explanation acts as a message and each masked evaluation is a channel use. Within this framework, the complexity of the explanation is captured by the entropy of the hypothesis class, while the query interface supplies information at a rate determined by an identification capacity per query. We derive a strong converse showing that, if the explanation rate exceeds this capacity, the probability of exact recovery necessarily converges to one in error for any sequence of explainers and decoders. We also prove an achievability result establishing that a sparse maximum-likelihood decoder attains reliable recovery when the rate lies below capacity. A Monte Carlo estimator of mutual information yields a non-asymptotic query benchmark that we use to compare optimal decoding with Lasso- and OLS-based procedures that mirror LIME and KernelSHAP. Experiments reveal a range of query budgets where information theory permits reliable explanations but standard convex surrogates still fail. Finally, we interpret super-pixel resolution and tokenization for neural language models as a source-coding choice that sets the entropy of the explanation and show how Gaussian noise and nonlinear curvature degrade the query channel, induce waterfall and error-floor behavior, and render high-resolution explanations unattainable.

23.
arXiv (CS.CL) 2026-06-16

Can LLM Coding Agents Reason About Time Series?

Large language models (LLMs) are increasingly being used for automated decision-making systems in finance, healthcare, or environmental monitoring. Time series data are ubiquitous in these fields, yet hard to process automatically. Can time series be analyzed by LLM agents? We examine three approaches: providing the agent with raw numerical data, using the LLM as a coding agent, or a combination of both. In the coding agent setup, the model iteratively queries the data using Python code. Using two time series understanding benchmarks, we show that agents with code access can outperform models processing raw data by up to 10%. However, even the best performing agent still answers about 22-34% of the questions incorrectly. To get insights into models' strategies and reasoning gaps, we analyze the model outputs with a strong LLM judge. Our analysis reveals that coding agents can select appropriate statistical tests, but often miss important nuances. Meanwhile, models with access to raw data can reach the right conclusions using back-of-the-envelope calculations.

24.
bioRxiv (Bioinfo) 2026-06-23

Comorbidity structure as an inductive bias: Comparing output-head designs for multi-label prediction of diabetes and myocardial infarction complications

Background: Clinical complications are often predicted with separate sigmoid outputs, even when the target labels arise from related pathophysiological processes. This paper asks whether output-layer choice should reflect both predictive convenience and the biological structure assumed among complications. The central premise is that label-dependence mechanisms are explicit hypotheses about comorbidity, not generic modelling additions. Methods: Output-head assumptions were compared across two clinically distinct multi-label prediction tasks. In Type 2 diabetes (T2D), six heads were evaluated for nephropathy, neuropathy, and retinopathy: independent baseline, linear additive, multiplicative, symmetric conditional random field (CRF), residual multilayer perceptron (MLP), and combined additive-multiplicative. In myocardial infarction (MI), four heads were evaluated for ventricular tachycardia, ventricular fibrillation, and atrioventricular block: independent baseline, linear additive, multiplicative, and symmetric CRF. All experiments used five training data fractions and seven independent seeds, with the same shared-backbone protocol within each disease setting. Results: In T2D, the symmetric CRF gave the most consistent improvement pattern, ranking highest at full data and at the two lowest data fractions while adding only three interaction parameters. At 20% training data, it was the only interaction head whose aggregate mean exceeded the independent baseline. The residual MLP, despite 123 interaction parameters, remained below the baseline across all T2D fractions. In MI, rankings changed across fractions: the multiplicative head led at 80% and 60%, the CRF led at 100% and 20%, and the baseline led at 40%. The combined additive-multiplicative head did not improve robustness in T2D and showed the largest negative baseline-relative deviations at lower fractions. Conclusions: The findings support a biology-guided view of output-layer design. A small constrained mechanism was most useful when its symmetry matched the shared microvascular structure of T2D, whereas the heterogeneous electrophysiology of MI produced no stable winner. Output-layer choice should therefore be reported and defended as an assumption about disease structure instead of a routine hyperparameter decision.

25.
medRxiv (Medicine) 2026-06-15

A controlled human infection model for symptomatic pertussis in North America using the pertactin-producing clinical isolate D420

Background Despite widespread vaccination, pertussis remains a poorly controlled disease globally and results in substantial annual morbidity and mortality, particularly in young children. Controlled human infection models (CHIMs) using the causative agent Bordetella pertussis are promising systems to enable the study of pertussis disease pathogenesis and immunology and to rapidly assess vaccines and therapeutics. While a pertussis CHIM that produces asymptomatic infection has been established in Europe, the development of a CHIM that leads to symptomatic illness would be advantageous for evaluating vaccine efficacy against both infection and disease. Methods Healthy participants 18-40 years of age were inoculated intranasally with one of eight doses (ranging from 104 to 108 colony forming units (CFU)) of the pertactin-producing B. pertussis isolate D420 at the challenge facility within the Canadian Center for Vaccinology (Nova Scotia, Canada). The study occurred in two stages. In stage one, the B. pertussis dose was escalated in cohort groups of five to six participants until reaching an endpoint where 70-90% of participants exhibited mild (non-severe, Grade 1 or 2) symptomatic infection, defined as the Human Infectious Dose 70-90 (HID70-90). In stage two, additional challenges were conducted for doses below, at, and above the identified HID70-90 to characterize the emerging pertussis model. For all challenge doses, participants were closely monitored during an inpatient stay of up to 24 days and post-discharge for laboratory-confirmed infection, pertussis symptoms, safety, and IgG antibody responses to four B. pertussis antigens including pertussis toxin, filamentous hemagglutinin, fimbriae, and pertactin. All participants received a five-day course of azithromycin, where timing of initiation depended on B. pertussis testing and symptoms. The study was conducted between July 4, 2022 and March 19, 2025. Findings Seventy-five participants were inoculated with one of the eight B. pertussis D420 challenge doses and completed the inpatient stay. From the stage-one dose escalation, we found that 107 CFU of B. pertussis D420 was the lowest dose that achieved the HID70-90, where 9 of 12 participants (75.0%) exhibited mild symptomatic infection. Following stage-two challenges, 16 of 22 total participants at 107 CFU (72.7%) developed mild symptomatic infection, thus verifying the HID70-90. The symptomatic infection rate below the HID70-90 at 5x106 CFU of D420 was 20.0% and above the HID70-90 at 5x107 and 108 CFU were 58.3% and 55.6%, respectively. Symptoms with elevated frequency for symptomatic infection (relative to background symptoms in non-infected) included nasal congestion, runny nose, fatigue, malaise, and cough. At the HID70-90, 50% of symptomatic infections included cough. Serological analyses of the four highest (stage-two) challenge doses (5x106, 107, 5x107, 108 CFU) revealed that antibody titres increased over time post-challenge. Seroconversion for at least one of the four studied antibodies was nearly twice as common for symptomatic (70.0%) than asymptomatic (35.7%) infection and was absent (0%) for non-infected. All infections were cleared following azithromycin treatment (100%) and there were no study-related serious adverse events. Interpretation A safe and reproducible symptomatic pertussis CHIM was achieved, providing a model for research on pertussis disease pathogenesis and immunology and for assessing vaccines and therapeutics. (Clinicaltrials.gov, NCT05136599).