Academic Intelligence · Curated Daily

探索全球前沿学术脉络

AcademicHub 汇聚顶级期刊与预印本平台的实时文献。定制您的专属科研雷达,利用大语言模型自动生成交叉领域文献分析简报。

01.
Nature (Science) 2026-06-17

Optical fibre gripper for high-performance 3D micromanipulation

作者:

Optical tweezers offer precise, non-contact control, but operate in a limited force regime and impose strict requirements on the characteristics of the targets as well as the environmental conditions1–4. Millimetre-scale mechanical tweezers can offer higher gripping force but are not suitable for precise manipulations5–11. Integrating microgrippers directly at the optical fibres provides a new approach for precise micromanipulation. However, existing fibre-integrated tweezers still face challenges in achieving high-performance manipulation of micro-objects (for example, single cells) within narrow spaces, mainly due to simplified architectures, constrained designs and millimetre-scale footprints12–14. Here we report a three-dimensional (3D) optical fibre gripper (OFG), which is fabricated by two-step, two-photon polymerization. The OFG consists of rigid photoresist microclaws and soft thermoresponsive hydrogel muscle doped with silver nanoparticles, and its size is only 38 × 38 × 61 μm3. The OFG exhibits a force-to-mass ratio of about 340 μN mg−1, outperforming previously reported fibre-integrated tweezers by one to two orders of magnitude. The OFG can manipulate opaque particles, irregular micromechanical components and diverse single-cell types. We further demonstrated its potential in 3D microassembly of complex microdevices (bearings, shafts and gearboxes) and biomimetic sampling in the narrow environment (<300 μm). These results position the OFG as a compact fibre-tip manipulator for 3D micromanipulation, offering reversible and tunable gripping in an intermediate force regime between optical field trapping and millimetre-scale mechanical tweezers. A miniature three-dimensional optical fibre gripper enables powerful, precise micromanipulation of particles and single cells in confined spaces, bridging the gap between optical and mechanical tweezers.

02.
arXiv (CS.CL) 2026-06-17

Learning task-specific subspaces via interventional post-training of speech foundation models

Speech foundation models, pre-trained on large corpora of unlabelled speech data, produce general-purpose representations which are useful across tasks. However, these representations encode information about salient speech variables in a distributed manner, while downstream speech tasks rely on only some of this variability. In this work, we propose a post-training refinement approach using interventional contrastive learning. By leveraging an interventional dataset and multi-part contrastive loss, we learn a transformation from the entangled representation space of speech foundation models into separate content and speaker subspaces. We evaluate the learnt representations on speaker verification and keyword spotting tasks, showing improved out-of-domain speaker verification performance and evidence that speaker and content information are separated across the learned subspaces.

03.
medRxiv (Medicine) 2026-06-11

Conversational Speech for Respiratory Triage in Primary Care: A Pilot Study

作者:

Background. Respiratory complaints account for a substantial share of adult ambulatory care visits, and triaging them accurately has direct consequences for antibiotic stewardship and pathogen-specific therapy. Prior work has investigated voice as a triage signal, but that literature is dominated by single-condition detection from scripted speech in crowdsourced or controlled clinical settings and has not been evaluated at primary care scale on conversational ambient audio. Methods. A dataset of 514,377 ambient-recorded primary care visits from 379,225 adult patients at a US clinic network was used, with per-visit clinically assigned ICD-10 diagnosis codes and de-identified demographic and geographic metadata. Patient audio was extracted from each doctor-patient conversation, and spectral, voice quality, and prosodic features were computed. Eleven binary classification tasks were defined, aligned with a respiratory triage cascade (e.g., acute respiratory versus acute non-respiratory illness, and lower versus upper respiratory tract infection). An acoustic model (feed-forward network) was trained independently for each task using patient-stratified five-fold cross-validation and evaluated on a held-out test set. Each task's model was also compared against six non-acoustic baselines using a single demographic, geographic, or temporal variable. The 11 trained classifiers were composed into a hierarchical cascade and illustrated as case studies on selected patients. Results. Test-set AUC across the 11 tasks ranged from 0.602 (95% CI: 0.588-0.614) to 0.745 (95% CI: 0.742-0.748), with a mean expected calibration error of 0.018. Six of eleven binaries outperformed all confounder baselines. Four binaries showed median within-stratum AUC of 0.62-0.70 when the confounder was held fixed, indicating acoustic discrimination beyond what the confounder alone explains. The exception was the pneumonia versus non-pneumonia lower respiratory tract infection binary, which failed against the patient-city confounder baseline, plausibly reflecting a clinic-level difference in ICD-10 coding. Conclusion. Conversational primary care audio carries acoustic signal that discriminates clinically meaningful respiratory contrasts. Absolute performance is moderate, but the conditions are stricter than prior work: conversational speech and differential-diagnosis contrasts among sick patients. This pilot study is a baseline for voice-based clinical AI moving beyond sick-versus-healthy detection toward differential-diagnosis panels and a proof-of-concept for hierarchical reasoning.

04.
arXiv (CS.LG) 2026-06-16

Empirical Study of Pop and Jazz Mix Ratios for Genre-Adaptive Chord Generation

作者:

arXiv:2605.04998v2 Announce Type: replace-cross Abstract: This revision updates a pop-to-jazz chord-generation rehearsal study. Best-epoch metrics still show that modest pop rehearsal preserves pop accuracy while improving jazz prediction, but v2 corrects released-checkpoint selection: the released F1 equals Phase 0, F2 had a transcription error, and ft-pop80-v2 restores a hash-distinct jazz-adapted F1 across 3 seeds.

05.
arXiv (CS.AI) 2026-06-16

Beyond Classification: A Cough Regression Benchmark for Respiratory Acoustic Foundation Models

arXiv:2606.15436v1 Announce Type: cross Abstract: Respiratory acoustic foundation models (FMs) excel at cough classification, yet their ability to predict continuous health quantities from cough audio remains largely unexplored, despite the clinical value of passive age, BMI, and disease probability estimation in settings where physical measurements are unavailable. We introduce the multi-model, multi-target cough regression benchmark evaluating five FMs (OPERA-CT, OPERA-CE, OPERA-GT, HeAR, M2D+Resp) across six targets on three datasets under subject-disjoint protocols, comparing linear, MLP-small, and full MLP regression heads. MLP-small beats the mean-predictor baseline on all tasks and linear probing in 23 of 30 model x task cases, with full MLP overfitting on small clinical data but recovering on larger sets, revealing a dataset size x head-capacity trade-off. HeAR leads within-dataset age regression on Coswara (9.12 yr MAE); its CIDRZ result is excluded from headline claims owing to possible HeAR-CIDRZ pretraining overlap. OPERA-GT is favored over OPERA-CT on age in all three datasets, with the CIDRZ margin within seed variance, extending a generative-pretraining advantage from breath to cough. HeAR and M2D+Resp reach near-full performance at N = 50 samples while OPERA models require N = 400. Cross-dataset transfer is strongly asymmetric as large diverse data generalises to small clinical populations (CoughVID to CIDRZ: -0.17 yr) but not vice versa (CIDRZ to Coswara: +2.43 yr, +26.6%).

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

Evaluating and Combating the Impact of Concept Drift on the Performance of Machine Learning-Based Phishing Detection Systems

arXiv:2606.11471v1 Announce Type: cross Abstract: The expansion of the digital domain has resulted in a substantial increase in digital communication, with email emerging as one of the most prominent channels. The proliferation of email communication is apparent in both professional and personal contexts, thereby creating numerous vulnerabilities for malicious actors to exploit. Spam emails, a form of unsolicited correspondence often bearing malicious intent towards recipients, have been an ongoing challenge for email users since the inception of email technology, and this problem has been exacerbated by the growth of the digital landscape. Email spam filters are integral components of email clients, engineered to identify potentially harmful messages and alert users to their malicious content. Phishing, frequently the initial phase of malware-based attacks, is evolving rapidly, with malware becoming increasingly sophisticated over time. A widely adopted approach for detecting malicious activity within malware and spam domains is the application of machine learning. Our aim is to assess the impact of the evolution within the spam email domain on these machine learning-based detection systems and to explore strategies for mitigating associated performance degradation.

07.
arXiv (CS.CL) 2026-06-16

When the Chain of Thought Knows Better: Failure Modes in Multi-Turn Reasoning Models

Failures in multi-turn reasoning models are largely invisible to terminal-score evaluation. A model can lock onto an unsafe stance early in a long dialogue, yet its final-turn refusal rate may appear indistinguishable from a robustly aligned baseline. To expose these hidden temporal dynamics, we propose a trace-level diagnostic - the CoT-Output 2x2 safety matrix. This framework labels every turn along two independent axes (internal reasoning and visible output), yielding four operationally defined failure cells: robust alignment, alignment faking, overt jailbreak, and a distinct failure mode we term context-injection failure (where the CoT maintains safe reasoning, but the visible output produces harm, highlighting a multi-turn manifestation of reasoning unfaithfulness). We evaluate three distilled reasoning targets against a fixed attacker across five oversight conditions, collecting 6750 turn-level observations on the Information-Hazard scenario. Our analysis reveals two reproducible vulnerabilities: an oversight paradox where explicit monitoring cues paradoxically increase alignment-faking rates rather than suppress them, and a context-injection failure where models lock onto unsafe external outputs despite safe internal states. We release the full dataset of multi-turn dialogues and CoT traces to support follow-up trace-diagnostic research.

08.
arXiv (CS.CL) 2026-06-11

Debiasing Without Protected Attributes: Latent Concept Erasure from Textual Profiles

Most fairness research in NLP assumes direct access to protected attributes such as gender, race, or nationality. In practice, however, such information is often unavailable due to privacy constraints, missing metadata, or legal restrictions, even though models may infer it from indirect textual cues. This raises a key question: can debiasing succeed without direct access to sensitive attributes? We propose H-SAL, which performs post-hoc concept and attribute erasure using self-description text as an implicit debiasing signal. To support this setting, we introduce a multi-domain Stack Exchange-based fairness benchmark for helpfulness prediction that includes both explicit and implicit signals, enabling comparison between standard debiasing with protected labels and debiasing without access to sensitive information. Across encoder and decoder-only language models, we find that implicit self-description often matches or outperforms explicit-label-based debiasing. Our results broaden representation-level fairness research and provide a new benchmark for studying debiasing under realistic data constraints.

09.
arXiv (math.PR) 2026-06-19

Maximal rigidity of random measure and uniqueness pairs: stealthy processes, quasicrystals and periodicity

arXiv:2512.10686v2 Announce Type: replace Abstract: This article investigates the phenomenon of maximal rigidity in spatial processes, where perfect interpolation of the process is possible from partial information, specifically, from its restriction to a strict subdomain, often resulting in a trivial tail $\sigma$algebra. A classical example known since the 1930's is that a time series is fully determined by its values on the negative integers if its spectrum has a gap, or at least a sufficiently deep zero. We extend such results to higher dimensions and continuous settings by establishing a connection with the concept of uniqueness pairs, rooted in the uncertainty principle of harmonic analysis. We present several other manifestations of this principle, unify and strengthen seemingly unrelated results across different models: quasicrystals and stealthy processes are shown to be maximally rigid on cones, and discrete integer-valued processes are necessarily periodic when they have a simply connected spectrum. Finally, we identify a surprising class of continuous fields with seemingly standard behavior, such as linear variance and finite dependency range, that undergo a phase transition: they are perfectly interpolable on B(0, $\rho$) for $\rho$ ___ 2 $\pi$ but exhibit no rigidity for $\rho$ > 2.

10.
arXiv (CS.AI) 2026-06-18

ARIADNE: Agnostic Routing for Inference-time Adapter DyNamic sElection

arXiv:2606.19079v1 Announce Type: new Abstract: The increasing deployment of parameter-efficient fine-tuning (PEFT) has led to model ecosystems in which a single backbone is paired with many task-specialized adapters. In this setting, inference-time queries often arrive without task labels, requiring the system to automatically select the most appropriate adapter from a growing and heterogeneous adapter pool. Existing routing methods either depend on access to adapter internals, such as weight decompositions or gradient-based statistics, or require additional router training, which limits scalability and portability as new adapters are added. We introduce ARIADNE, a training-free, adapter-agnostic routing framework for dynamic adapter selection at inference time. ARIADNE represents each adapter through a set of centroids computed from embeddings of its training set, capturing the data distribution associated with that adapter. Given an unlabeled input, it selects an adapter by measuring proximity to these centroids in latent space. Because routing is performed entirely in the input embedding space, ARIADNE is compatible with arbitrary PEFT methods and requires no modification to the adapters or training procedures. Primarily evaluated with Llama 3.2 1B Instruct on 23 diverse NLP tasks, ARIADNE recovers 97.44% of the upper bound performance. Scaling to 44 tasks, it achieves 89.7% average selection accuracy, without additional training or access to adapter internals.

11.
arXiv (CS.LG) 2026-06-12

Allocating Human Oversight in AI-Enabled Analytics

arXiv:2604.12497v2 Announce Type: replace Abstract: Organizations increasingly deploy AI as a low-cost prediction layer in customer-facing decision processes, including demand sensing, service-quality monitoring, product testing, and market research, but AI-generated signals are unevenly reliable across tasks, products, and customer segments. Firms therefore still need scarce human validation (labels, audits, survey responses, or follow-up measurements) to anchor AI outputs to ground truth. Because human ground truth is itself noisy, varying across labelers and even across repeated judgments, the firm must collect and average several human labels per task, which makes human validation costly. We study how to allocate a limited human-validation budget across many AI-assisted tasks when reliability is heterogeneous and unknown before deployment. We cast this within tuned prediction-powered inference. Each human label both sharpens the AI-assisted estimate and reveals the task's rectification difficulty, the variance that remains after the AI prediction is optimally used as a control variate. If difficulties were known, the optimal allocation would follow a Neyman square-root rule; because they are unknown, we propose a policy based on upper confidence bounds that learns them online and steers validation toward tasks where AI is least reliable. We prove that the policy's terminal efficiency loss relative to the oracle allocation vanishes as the budget grows. In synthetic experiments and a real digital-twin survey with 68 tasks and over 2000 respondents, it closes most of the gap to the oracle when reliability is heterogeneous, outperforming uniform and epsilon-greedy allocation; on the survey data it also outperforms explore-then-commit pilot designs and cuts uniform's 10–12% gap to 2–6%. The value of AI depends not only on model accuracy but also on the operational policy that targets human oversight where AI errors matter most.

12.
arXiv (CS.LG) 2026-06-18

EfficientRollout: System-Aware Self-Speculative Decoding for RL Rollouts

arXiv:2606.18967v1 Announce Type: new Abstract: Reinforcement learning (RL) has become a representative post-training paradigm for LLMs, enabling strong reasoning and agentic capabilities. However, rollout generation remains a dominant latency bottleneck because autoregressive sampling decodes responses sequentially and a small number of long-tailed generations often determine completion time. Speculative decoding (SD) offers a natural way to address this bottleneck, as it is a well-established technique for serving fixed LLMs that reduces latency by rapidly drafting tokens and accepting them through parallel verification while preserving the target-model distribution. However, its practical speedups do not directly carry over to RL rollouts: (i) the evolving target policy makes any fixed drafter increasingly mismatched with the policy's output distribution; and (ii) active batch sizes shrink throughout rollout decoding, shifting decoding from compute-bound to memory-bound regimes where parallel verification can exploit underutilized compute. Therefore, accelerating RL rollouts requires both a drafter that remains effective under long, high-temperature generations from an evolving policy and system-aware use of SD that avoids compute-bound regimes. We present EfficientRollout, a system-aware self-SD framework designed to address this gap for RL rollouts. EfficientRollout induces a quantized drafter from the target model (i.e. self-speculative decoding), keeping it coupled to the evolving policy without separate drafter pretraining or online adaptation. It further coordinates a system-aware SD toggle policy with acceptance-aware draft-length adaptation, enabling speculation only in beneficial regimes while matching the drafting budget to evolving drafter quality. EfficientRollout reduces rollout and end-to-end latency by up to 19.6% and 12.7%, respectively, over an accelerated AR rollout baseline, while preserving final model quality.

13.
arXiv (CS.CL) 2026-06-12

Does AI Reviewer See the Full Picture? Attacking and Defending Multimodal Peer Review

The integration of Large Language Models (LLMs) and Multimodal LLMs (MLLMs) into scientific peer-review workflows introduces novel and significant risks for adversarial manipulation, especially given the multimodal nature of scientific papers where figures, not just text, convey core evidence. This creates a significant gap: current robustness studies on AI peer-review are overwhelmingly text-only. Moreover, the problem is distinct from standard jailbreaking, as a peer-review attack seeks to induce a domain-specific, targeted failure (e.g., "inflate this score") rather than a general safety policy violation, for which no practical defenses exist. To address this, we introduce PaperGuard, the first comprehensive benchmark designed to systematically evaluate and defend AI-generated peer-review against these domain-specific, cross-modal attacks. Our framework is built on three pillars: (1) a new multimodal peer-review dataset spanning multiple scientific domains; (2) a unified suite of attacks, including black-box prompt injections and white-box perturbations, specifically designed to target both text (GCG) and figures (PGD); and (3) a practical defense, motivated by the long-context challenge of academic papers, that uses chunk-based embedding search to efficiently localize and mitigate harmful instructions. Our extensive experiments, conducted across state-of-the-art models, confirm that AI reviewers are pervasively vulnerable. PaperGuard establishes the foundational benchmark, protocols, and actionable defense necessary to pioneer trustworthy, attack-resilient AI-assisted scholarly reviewing.

14.
arXiv (CS.CV) 2026-06-12

VISTA: An End-to-End Benchmark for Visual Spec-to-Web-App Coding Agents

We present VISTA (VIsual Spec-To-App Benchmark), a benchmark for evaluating the end-to-end web-app generation capabilities of LLM-based agents. Unlike prior code generation benchmarks that focus on algorithmic tasks, VISTA targets realistic UI-centric development, where agents must produce functional, visually coherent applications from underspecified inputs. We define five prompt-information conditions that vary along two axes, visual/structural fidelity and stack constraint: (1) text only with free stack choice, (2) text with reference screenshots under three specified stacks, (3) text with reference screenshots under free stack choice, (4) text with screenshots and pruned Figma structure under a single specified stack, and (5) text with screenshots and pruned Figma structure under free stack choice. To enable robust evaluation, each page in the benchmark is manually annotated with interactive UI components and around three visual anchor points, addressing the well-known limitations of script-based testing tools such as Playwright in open-ended code generation settings. Evaluation combines DOM-grounded reference matching, behavior-specific browser tests, and CLIP-based visual similarity, jointly measuring structural alignment, behavioral completeness, and overall visual fidelity. We use VISTA to assess four agent systems drawn from two model families and two harnesses, finding that visual fidelity and functional correctness are partially decoupled across both input conditions and agents, and that agent editing style varies sharply but is largely orthogonal to task quality. VISTA establishes a rigorous and reproducible foundation for advancing agent-based software engineering research.

15.
arXiv (CS.CV) 2026-06-16

KeepLoRA++: Continual Learning with Layer-Scaled Residual Gradient Adaptation

Continual learning for pre-trained vision-language models requires balancing three competing objectives: retaining pre-trained knowledge, preserving knowledge from a sequence of learned tasks, and maintaining the plasticity to acquire new knowledge. This paper presents KeepLoRA++, balancing these objectives through a unified dual-dimensional knowledge retention mechanism. We analyze knowledge distribution of Transformer architecture from both inter-layer and intra-layer perspectives. The inter-layer perspective examines how retention is distributed across layers, while the intra-layer perspective focuses on the parameter space within each layer. Our analysis reveals a structural property: general transferable knowledge is mainly encoded in the shallow layers and the principal subspace of the parameters, while task-specific adaptations are localized in the deep layers and the residual subspace. Motivated by this insight, KeepLoRA++ introduces a layer-scaled residual gradient adaptation method. New tasks are learned by restricting LoRA parameter updates to the residual subspace, combined with a shallow-to-deep layer scaling, to prevent interference with previously acquired capabilities. Specifically, the gradient of a new task is projected onto a subspace orthogonal to both the principal subspace of the pre-trained model and the dominant directions of previous task features, while simultaneously assigning smaller update magnitudes to shallow layers and larger ones to deeper layers. Our theoretical analysis and empirical evaluations confirm that KeepLoRA++ successfully balances these three competing objectives, consistently outperforming representative baselines across image classification, visual question answering, and video understanding tasks.

16.
arXiv (math.PR) 2026-06-19

The t-Split Two-Periodic Aztec Diamond Model

arXiv:2606.19507v1 Announce Type: new Abstract: In this work we consider an Aztec diamond model split into two unequal regions which are asymptotically fixed in size. Each region is weighted with a distinct two-periodic weighting. We refer to this model as the t-split two-periodic Aztec diamond, to signify its difference from the previous work title Split Two-Periodic Aztec Diamond, where the model was split into two equal regions. We derive an integral expression for the correlation kernel of the model and give a partial description of the scaling limit behavior, along with a conjecture for the remainder. We refer to the larger and smaller sides of the model as the dominant and non-dominant sides, and to the location of the weight change as the interface. The dominant side exhibits a limit shape that depends only on its own weighting and is identical to that of the two-periodic Aztec diamond, while the non-dominant side appears to have a novel limit shape that depends on both weightings and the location of the interface. Lastly, we consider the complete limit shape in the case where the dominant side two-periodic parameter goes to 0.

17.
arXiv (CS.CL) 2026-06-16

A Self Consistency Based Reranking for Narrative Question Answering

Narrative question answering (NQA) is a challenging task in natural language processing that requires models to understand long textual contexts, capture relationships across events, and generate coherent responses. Despite recent advances in pretrained language models, most existing approaches rely on a single decoding output during inference, making them sensitive to generation variability and often resulting in incomplete or inconsistent answers .To address this limitation, we propose a self-ensemble Self-Consistency-Based reranking framework for narrative question answering. The proposed method generates multiple candidate answers for each story-question pair and selects the final answer based on semantic agreement among the generated responses. This allows the model to explore diverse answer formulations while improving robustness through consensus-based selection without requiring modifications to the underlying architecture .The framework combines pretrained and fine-tuned language generation with multi-answer inference and similarity-based reranking. We evaluate the proposed approach on the NarrativeQA dataset using multiple models, including FLAN-T5 (Base and Small) and Pegasus-Large, under both baseline and fine-tuned settings .Experimental results demonstrate that the proposed method consistently improves performance across all models. In particular, FLAN-T5-Base achieves the best overall performance, improving from 82.32% to 86.66% (+4.34%) when combined with self-ensemble inference. Additionally, the largest improvement is observed with Pegasus-Large, which increases from 72.50% to 87.07% (+14.57%), highlighting the effectiveness of the proposed strategy.

18.
arXiv (CS.CV) 2026-06-12

Radar-Guided Polynomial Fitting for Metric Depth Estimation

We propose POLAR, a novel radar-guided depth estimation method that introduces polynomial fitting to efficiently transform scaleless depth predictions from pretrained monocular depth estimation (MDE) models into metric depth maps. Unlike existing approaches that rely on complex architectures or expensive sensors, our method is grounded in a fundamental insight: although MDE models often infer reasonable local depth structure within each object or local region, they may misalign these regions relative to one another, making a linear scale and shift (affine) transformation insufficient given three or more of these regions. To address this limitation, we use polynomial coefficients predicted from cheap, ubiquitous radar data to adaptively adjust predictions non-uniformly across depth ranges. In this way, POLAR generalizes beyond affine transformations and is able to correct such misalignments by introducing inflection points. Importantly, our polynomial fitting framework preserves structural consistency through a novel training objective that enforces local monotonicity via first-derivative regularization. POLAR achieves state-of-the-art performance across three datasets, outperforming existing methods by an average of 24.9% in MAE and 33.2% in RMSE, while also achieving state-of-the-art efficiency in terms of latency and computational cost.

19.
bioRxiv (Bioinfo) 2026-06-11

VFUSE: Virulent Feature Understanding with Sparse autoEncoders

Generative models have shown remarkable progress in a variety of domains such as protein design, but such power enables the opaque generation of hazardous proteins. In this work, we introduce VFUSE (Virulent Feature Understanding with Sparse autoEncoders), a mechanistic interpretability approach that trains SAEs on diffusion-transformer activations to audit protein models for hazard-aware features. We apply VFUSE to RoseTTAFold3 and RFDiffusion3, popular open-weight models for protein folding and synthesis. We find that for certain blocks, linear probes detect hazardous designs significantly better when fit in the SAE latent space over the original model's representations: improving interpretability without sacrificing model performance. Furthermore, we identify monosemantic features from the SAE that fire only on hazardous designs at up to AUROC 0.84 (q < 10-13).

20.
arXiv (CS.AI) 2026-06-19

Before the Pull Request: Mining Multi-Agent Coordination

arXiv:2606.19616v1 Announce Type: cross Abstract: Autonomous coding agents now open millions of pull requests, yet large-scale studies find their PRs are produced faster but accepted less often - a coordination and trust gap that pull-request-level telemetry cannot explain. We argue the missing signal lives before the PR, in how concurrent agents claim, divide, and collide over shared work. We study this process through grite, our open-source coordination substrate that needs no central server and stores its records inside git itself, so its append-only, signed event log captures the coordination process directly. We show that (i) this shared substrate reduces duplicate and conflicting work at bounded overhead - the share of work that merely re-does a teammate's task falls from 78% to 0% while useful throughput more than triples; (ii) every agent's copy of the log converges to the same state with no write silently dropped, where a file-based tracker loses concurrent writes; and (iii) the log is a mineable artefact from which concrete failure modes - conflicting edits, lock starvation, redundant rediscovery, race-to-close - are automatically recoverable with provenance, several invisible in pull-request history. We release the dataset, harness, and mining toolkit.

21.
medRxiv (Medicine) 2026-06-22

Referral pathways, ETAT triage acuity, and inpatient outcomes among children presenting to a national tertiary paediatric emergency unit in Ghana: a prospective cohort study

Emergency referral systems in sub-Saharan Africa are fragmented, and children reaching tertiary facilities through different referral pathways often arrive in advanced clinical states. Prospective data simultaneously characterising referral patterns, triage acuity at presentation, diagnostic case mix, and inpatient mortality at a national tertiary paediatric emergency unit are lacking from West Africa. This prospective cohort study enrolled 675 consecutively presenting children aged one month to 12 years at the Paediatric Emergency Unit of Korle Bu Teaching Hospital, Accra, Ghana, from February to December 2019. The primary outcome was all-cause inpatient mortality. Key variables collected included referral status and facility tier, Emergency Triage Assessment and Treatment (ETAT) triage category, ICD-10 diagnostic classification, Oyedeji socioeconomic classification, and time from symptom onset to PEU registration. Crude odds ratios were computed for all candidate predictors. Multivariable logistic regression was conducted using complete case analysis (n = 613). Of 675 children, 63.0% (n = 425) were referred from another health facility; referred children had higher ETAT emergency triage category rates than self-presenting children (32.7% vs 27.6%, p < 0.001). Overall inpatient mortality was 9.9% (67/675). Mortality varied by referral source: 16.7% among secondary/regional hospital referrals, 11.0% among lower-tier facility referrals (district, municipal, CHAG, polyclinic, private, health centre, and maternity home facilities combined, n = 356), 7.6% among self-presenting children, and 7.4% among tertiary referrals. Overall, 30.8% of children were classified as ETAT emergencies on arrival, with case fatility rate of 21.6%. The three most common diagnostic domains were respiratory conditions (17.2%), blood and haematological disorders (17.0%), and digestive presentations (16.4%). Inpatient mortality was highest in neoplastic disease (33.3%, n = 30) and circulatory presentations (31.0%, n = 29). In the primary multivariable analysis (n = 613, 51 events; events-per-variable ratio 4.2), no referral tier was independently associated with inpatient mortality after adjustment. Referral from secondary/regional hospitals showed a borderline non-significant association (adjusted odds ratio 3.09, 95% CI 0.96 to 9.90, p = 0.058). School going children (60-119 months) had higher odds of inpatient death than infants (adjusted odds ratio 5.56, 95% CI 1.16 to 26.53, p = 0.032), as did adolescents (adjusted odds ratio 10.01, 95% CI 2.15 to 46.69, p = 0.003). ETAT emergency category and lower socioeconomic status were not independently significant in this model. A pre-specified sensitivity analysis using the full analytic cohort (n = 674, events-per-variable ratio 6.7) with collapsed referral categories did not confirm any referral tier association; ETAT emergency category and lower SES were independently associated in the sensitivity model. All multivariable estimates should be regarded as exploratory. This prospective cohort provides simultaneous characterisation of referral patterns, ETAT triage acuity, diagnostic case mix, and inpatient mortality at a national tertiary paediatric emergency unit in West Africa. The referral-mortality gradient and high ETAT emergency category proportion document the severity of illness arriving through different referral pathways at this facility. The association between secondary/regional hospital referral and inpatient mortality is hypothesis-generating and requires replication in an adequately powered multicentre study before any service-level conclusions can be drawn.

22.
arXiv (CS.AI) 2026-06-18

FoMoE: Breaking the Full-Replica Barrier with a Federation of MoEs

arXiv:2606.19025v1 Announce Type: cross Abstract: Pre-training Large Language Models (LLMs) typically demands large-scale infrastructure with tightly coupled hardware accelerators. While increasing model and dataset scale remains the dominant driver of performance, Mixture-of-Experts (MoEs) architectures have recently achieved state-of-the-art results by decoupling parameter count from computational cost. This efficiency enables training massive models on constrained compute budgets, yet it typically requires the high-speed interconnects of a single datacenter. To overcome these physical limits, recent approaches such as DiLoCo and Photon use low-communication data-parallel methods to enable scaling across geographically distributed, weakly connected data centers. However, these methods suffer from a fundamental inefficiency: they require full model replicas at every site, which imposes prohibitive memory constraints and communication overheads. In this work, we introduce FoMoE, a system that breaks the full-replica paradigm by partitioning expert layers across workers. We demonstrate that FoMoE: (I) reduces communication costs by up to 1.42x over efficient baselines and 45.44x over DDP via partial expert replication in the studied regimes; (II) achieves empirical throughput speedups of up to 1.4x through a novel skip-token mechanism; and (III) shows stable routing in the trained proxy regimes and projects the communication/memory benefits to 100B-scale configurations through system modelling.

23.
arXiv (CS.CL) 2026-06-18

LLMs Struggle to Measure What Distinguishes Students of Different Proficiency Levels: A Study of Item Discrimination in Reading Comprehension Assessment

Item discrimination is a fundamental psychometric property of educational assessment, which measures whether an item meaningfully distinguishes students with higher proficiency from students with lower proficiency. While various existing works have explored whether large language models (LLMs) can estimate item difficulty, it remains unclear whether they can capture item discrimination. In this work, we evaluate 42 proprietary and open-weight LLMs in zero-shot settings using two complementary approaches: direct discrimination prediction, where models explicitly estimate an item's discrimination value from its content, and response-based Classical Test Theory (CTT) calibration, where LLM answers are treated as synthetic student responses to compute discrimination scores. Our results show that direct prediction yields weak alignment with human-calibrated discrimination: the best-performing model reaches only a Spearman correlation of 0.152. Response-based CTT calibration provides a stronger but still limited signal, with the all-persona synthetic respondent pool reaching a Spearman correlation of 0.241. These findings highlight item discrimination as an open challenge for LLM-based psychometric evaluation: current LLMs contain non-random discrimination-relevant signal, but they do not yet reliably capture how assessment items distinguish human students.

24.
arXiv (CS.AI) 2026-06-19

DynAMO:Dynamic Asset Management Orchestration via Topological Multi-Agent Scheduling

arXiv:2606.19382v1 Announce Type: cross Abstract: While LLM-powered agents offer end-to-end automation for industrial asset lifecycles, real-world Industry 4.0 deployment is hindered by latency, concurrency instability, and safety risks. We present DynAMO (Dynamic Asset Management Orchestration), a deployment-ready engine using a Plan-then-Execute architecture to generate verifiable workflow graphs. DynAMO supports both SequentialWorkflow (topological execution) and ParallelWorkflow (dependency-aware concurrency). By dynamically identifying independent tasks, DynAMO preserves structural correctness and safety while significantly improving efficiency through controlled reasoning overlap. Across six controlled experiments on the AssetOpsBench industrial benchmark, DynAMO demonstrates substantial performance and robustness gains. Parallel execution reduces end-to-end latency by a median of 1.6x over sequential orchestration, rising to 1.8x on highly parallelizable workflows. After instrumenting external tool calls with realistic latencies, a latency decomposition shows that LLM reasoning and orchestration still account for more than 90% of execution time, identifying model inference as the primary system bottleneck. Structured context pruning reduces inference latency by approximately 30%, and DynAMO maintains correct functional behaviour (task completion, agent sequencing, and output quality) while exhibiting graceful degradation under controlled fault injection. Reproducibility analysis further confirms stable execution under repeated runs, with parallel scheduling reducing latency variance. These findings establish DynAMO as a practical blueprint for scalable, safe, and latency-aware agent deployment in Industry 4.0 automation pipelines. Code is available at: https://github.com/kushwaha001/DynAMO

25.
arXiv (CS.LG) 2026-06-12

Normative Robustness as a Frontier for Non-Verifiable Reasoning in LLMs

arXiv:2606.12731v1 Announce Type: new Abstract: As LLMs increasingly serve in advisory and deliberative roles, users rely on them for non-verifiable reasoning in domains lacking objective ground truths. However, traditional evaluations of LLM reasoning focus almost exclusively on fact-based domains, such as mathematics and science, leaving uncertainty over whether and to what degree models can handle ambiguous, subjective, or value-laden problems over time. To address this concern, we propose moral reasoning as a paradigmatic subdomain of non-verifiable reasoning. We define moral robustness as a model's capacity to exhibit sound moral reasoning across time and contexts, and we introduce a scalable, adversarial, multi-turn evaluation framework to empirically measure this capability. We simulate 48,000 user-agent moral deliberations across four frontier LLMs, varying premise relevance, premise order, conversation duration, and the user's stated moral view. We find that models successfully ignore morally-irrelevant distractors, but shift their reasoning by up to 6.5%, on average, towards the user's stated preferred moral view, and varying their reasoning depending on factors such as order (altering moral judgments by order in 13-22% of the cases) and duration (altering moral judgments between single-turn and multi-turn in 10-24% of the cases). Our analysis indicates that models tailor not just their final verdicts but their underlying justifications to align with a user's moral viewpoint - a failure mode we characterize as moral deliberative sycophancy.