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.
arXiv (quant-ph) 2026-06-24

Certifying Quantum Optimization and Circuit Cutting by Using Quantum-Classical Moment Duality

作者:

arXiv:2606.23727v1 Announce Type: new Abstract: We establish a direct quantum-classical duality based on the degree-$2$ Sum-of-Squares (SoS) semidefinite programming cone: the matrix of two-qubit Pauli-$Z$ correlation functions obtained from any quantum state $\rho$ is automatically a feasible point of the classical Goemans-Williamson (GW) relaxation. This observation provides a universal ``safety net'' for quantum optimization algorithms: applying GW random hyperplane rounding to the quantum-driven moment matrix yields a certified expected cut value $\mathbb{E}[\mathrm{Cut}] \ge \alpha_{\mathrm{GW}}\langle\mathcal{H}\rangle_\rho$, valid for every state produced by variational algorithms such as QAOA or the Variational Quantum Power Method (VQPM), regardless of convergence quality. We further show that the same moment matrix reveals the tensor-product structure of the underlying unitary circuit, enabling a polynomial-time, correlation-based circuit cutting procedure with rigorous error bounds. The framework is validated numerically on Max-Cut instances for variational quantum algorithms and on random states for circuit cutting, demonstrating that the cheap two-point correlation data are sufficient to locate near-optimal bipartitions and that the theoretical error bounds hold in practice.

02.
medRxiv (Medicine) 2026-06-24

Food insecurity, caloric intake and nutritional status among children under 5 years old: a predictive modelling analysis of the MAL-ED multi-country cohort

Background For children at risk of acute malnutrition, being able to predict and forecast dietary intakes and/or nutritional evolution would support decision-making, particularly in crisis settings where ground data collection is unfeasible or scant. We explored whether statistical models could offer accurate predictions of caloric intake or anthropometric (weight-for-height Z score, WHZ) changes, given intake, household food insecurity and other plausible predictors. Methods We reanalysed data from the Malnutrition and Enteric Disease (MAL-ED) multi-country (Bangladesh, Brazil, India, Nepal, Pakistan, Peru, South Africa, Tanzania) birth cohort (2009-2014), which consistently tracked household food insecurity experience, dietary intake, anthropometry, infectious disease symptoms, breastfeeding and other variables among children 9 to 35 months old. We quantified the performance on cross-validation of three models: (M1) change in WHZ as a function of household food insecurity; (M2) change in WHZ as a function of caloric intake; (M3) caloric intake as a function of household food insecurity. We compared random forests, lasso regressions, additive models and generalised boosted regressions. All models included age, sex, birth weight, urban versus rural residence, breastfeeding status and the longitudinal prevalence of diarrhoea, acute respiratory infection and fever as additional predictors. Results Altogether, M1, M2 and M3 leveraged 2957, 23,651 and 2013 longitudinal child observations, respectively. Both at country and individual level, there was low correlation among the key variables of interest. All three models featured low performance and moderate to extreme regression dilution, even when fitted to each country cohort separately. Discussion This secondary analysis based on data from a rigorous observational study suggests that statistical prediction of key variables along the causal pathway to childhood acute malnutrition may not be feasible. These negative findings may in part be explained by error in predictor measurement and the narrow range of both predictor and outcome values in the MAL-ED cohort, relative to the more extreme scenarios common to crisis settings. They also imply that mechanistic models requiring caloric intake as an input cannot rely on a statistical shortcut of this kind and must instead depend on empirical data or scenario assumptions.

03.
arXiv (math.PR) 2026-06-18

Kemeny's constant minimization for reversible Markov chains via structure-preserving perturbations

arXiv:2510.24679v4 Announce Type: replace-cross Abstract: Kemeny's constant measures the efficiency of a Markov chain in traversing its states. We investigate whether structure-preserving perturbations to the transition probabilities of a reversible Markov chain can improve its connectivity while maintaining a fixed stationary distribution. Although the minimum achievable value for Kemeny's constant can be estimated, the required perturbations may be infeasible. We reformulate the problem as an optimization task, focusing on solution existence and efficient algorithms, with an emphasis on the problem of minimizing Kemeny's constant under sparsity constraints.

04.
medRxiv (Medicine) 2026-06-22

Development of a Novel Risk Prediction Model for Rheumatoid Arthritis-Associated Interstitial Lung Disease (RA-ILD): A Longitudinal Study

Background: Interstitial lung disease (ILD) is one of the most common and potentially most devastating extra-articular complication of rheumatoid arthritis (RA) and is associated with substantial morbidity and mortality. However, reliable tools for the early identification of ILD in patients with RA remain limited. This study aimed to identify plasma protein biomarkers of RA-ILD and develop an interpretable machine learning model for risk prediction using data from the UK Biobank. Methods: We first evaluated the association between baseline RA and the risk of incident ILD in the UK Biobank using Cox proportional hazards models. Mendelian randomization analysis was then performed to investigate the potential causal relationship between RA and ILD. Finally, we analyzed 2,920 plasma proteins measured using the Olink platform in 781 eligible RA patients. Proteins associated with ILD risk were identified using Cox proportional hazards models and subsequently used to construct eight machine learning models. Model performance was assessed using the receiver operating characteristic curve (ROC) and decision curve analysis. The best-performing model was further interpreted using Shapley additive explanations (SHAP) to evaluate feature importance. Results: Compared with participants without RA, Patients with baseline RA had a significantly higher risk of developing ILD (Hazard ratio: 4.425, 95% CI: 3.549,5.518). The MR supported a potential causal association between RA and ILD (Odds ratio: 1.227, 95% CI: 1.121,1.343). Among the eight machine learning models, the CatBoost model showed the best performance, achieving an area under the curve (AUC) of 0.884 (95% CI: 0.773,0.996). The SHAP analysis identified LAG3, NPC2, and LAMP3 are the three most important plasma protein predictors of ILD development in patients with RA. Conclusion: Plasma proteomics combined with machine learning may provide a promising approach for identifying biomarkers and predicting ILD risk in patients with RA. LAG3, NPC2, and LAMP3 may serve as candidate biomarkers for RA-ILD and warrant further validation. Keywords: Rheumatoid arthritis, Interstitial lung disease, Mendelian randomization, Machine learning, Plasma proteins.

05.
arXiv (CS.CL) 2026-06-12

LLM-based Embeddings: Attention Values Encode Sentence Semantics Better Than Hidden States

Sentence representations are foundational to many Natural Language Processing (NLP) applications. While recent methods leverage Large Language Models (LLMs) to derive sentence representations, most rely on final-layer hidden states, which are optimized for next-token prediction and thus often fail to capture global, sentence-level semantics. This paper introduces a novel perspective, demonstrating that attention value vectors capture sentence semantics more effectively than hidden states. We propose Value Aggregation (VA), a simple method that pools token values across multiple layers and token indices. In a training-free setting, VA outperforms other LLM-based embeddings, even matches or surpasses the ensemble-based MetaEOL. Furthermore, we demonstrate that when paired with suitable prompts, the layer attention outputs can be interpreted as aligned weighted value vectors. Specifically, the attention scores of the last token function as the weights, while the output projection matrix ($W_O$) aligns these weighted value vectors with the common space of the LLM residual stream. This refined method, termed Aligned Weighted VA (AlignedWVA), achieves state-of-the-art performance among training-free LLM-based embeddings, outperforming the high-cost MetaEOL by a substantial margin. Finally, we highlight the potential of obtaining strong LLM embedding models through fine-tuning Value Aggregation.

06.
medRxiv (Medicine) 2026-06-22

Associations of Chemical Exposures with Psychological Distress and Depression Diagnosis among Waste Pickers in Brasilia, Brazil: A Cross-Sectional Study

Introduction: Waste pickers face chemical exposures. We evaluated whether chemical exposure is associated with psychological distress and depression. Methods: A 2017 cross-sectional survey included 1,141 waste pickers working in the Estrutural open dump in Brasilia, Brazil. Participants self-reported occupational exposure to 11 chemical categories, 17 psychological distress symptoms, and depression diagnoses. Associations of chemical exposure with mean psychological distress scores and depression prevalence were assessed, adjusted for age, sex, marital status, and income. Results: Mean psychological distress score was higher among those exposed to any chemical (mean of 8.1 vs 6.1; adjusted mean difference [aMD]: 1.8 [0.9, 2.7]) and higher among those exposed to each of 11 chemical categories, for example, smoke (aMD: 1.2 [0.6, 1.7]), batteries (aMD: 1.5 [1.0, 1.9], and oils (aMD: 1.3 [0.9, 1.8]). Depression was more prevalent among those exposed to oils (16.6% vs 10.6%; adjusted prevalence difference [aPD]: 6.3% [95% CI: 2.3, 10.2]), cleaning products (aPD: 5.4% [1.2, 9.5]), medications (aPD: 4.7% [0.6, 8.8]), and aerosols (aPD: 5.3% [1.3, 9.3]) but, not smoke, batteries, greases, insecticides, solvents, paints, chemical containers, or any chemical. Conclusion: These associations highlight the need to consider policy level protections for waste pickers to reduce chemical exposure and guard against psychological distress. Further research is necessary to explore which specific chemicals, within broad chemical categories, are associated with psychological distress and depression.

07.
arXiv (quant-ph) 2026-06-12

Candidate overtone shear horizontal SAW resonators in thin-film lithium niobate for intermodal acousto-optic modulation

arXiv:2606.12853v1 Announce Type: cross Abstract: The merits of thin-film surface acoustic wave (SAW) devices are pivotal to develop the high-performance intermodal acousto-optic modulators. In this work, we have proposed shear-horizontal (SH) SAW resonators for anticipated intermodal acousto-optic modulation on the thin-film lithium niobate platform. Through optimization of the cut angle of LN films, the SAW wavelength, and the thickness of interdigital transducer (IDT) electrodes, the calculated acousto-optic overlap factors utilizing SH0 modes are improved by more than an order of magnitude compared with those of Rayleigh modes. Furthermore, we have fabricated and characterized three kinds of proof-of-principle SH0 mode devices without/with grating reflectors. The electromechanical coupling coefficients (keff^2) and quality factors (Q) in the overtone resonators with grating reflectors are systematically evaluated, featuring the highest Q of 843 with the compromised keff^2 of 0.96%-4.72%. The results reveal that the temperature coefficients of frequency (TCF) of Rayleigh modes vary across various overtones, whereas the SH0 modes exhibit TCFs in the range of 32.3-68.9 ppm/C. Our fabricated SH0-mode overtone resonators demonstrate the capability of operating at power levels up to 29 dBm without electrode damage, offering a promising paradigm for robust and high-efficiency intermodal acousto-optic modulators with potential applications in integrated optical signal processing, microwave photonics,and quantum information technologies.

08.
arXiv (CS.AI) 2026-06-16

UXBench: Measuring the Actionability of LLM-Generated UX Critiques

arXiv:2606.16262v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly deployed as UX judges that inspect interfaces, diagnose usability problems, and propose repairs. Yet no controlled benchmark measures whether the resulting critiques are reliable and actionable across heterogeneous product surfaces. We introduce UXBench, a benchmark for evaluating LLMs as interaction-grounded UX judges. UXBench comprises local-first runnable web fixtures spanning ten product-surface families, paired with coverage-gated browser exploration that forces models to collect interaction evidence before reporting. Each judge model produces a structured UX report over seven rubric dimensions; report quality is measured by whether a fixed downstream repair agent can improve the interface based on the critique. We evaluate eight frontier models under both an automated repair-lift protocol and a blind human validation study. Results show that UX judging is neither saturated nor one dimensional: models differ meaningfully in report actionability, exhibit distinct rubric-level repair signatures, vary in fixture-level reliability, and trade leadership across surface categories

09.
medRxiv (Medicine) 2026-06-13

Projected population level impact and cost-effectiveness of clinic and community-based tuberculosis screening approaches

The South Africa National Department of Health have set ambitious targets to scale up TB testing, focusing primarily on clinic attendees. In the context of declining funding for TB care and prevention, the most cost-effective approaches for targeting testing should be identified. We developed a mathematical model of TB in South Africa, explicitly incorporating clinic attendance by sex and HIV/ART status. We simulated six screening approaches over 2026-2035 (individually and in combination): three clinic-based (symptom screening, intensified targeted universal TB testing [TUTT, symptom-agnostic sputum testing of clinic attendees in key risk groups], and intensified TUTT allowing saliva samples) and three targeted community-based (community radiographic screening, symptom screening, and universal Xpert Ultra testing), each implemented at a range of coverage levels. Model outputs were combined with a mechanistic cost function to estimate potential impact and cost-effectiveness from a societal perspective. The most cost-effective standalone approach was community radiographic screening at 10% annual population coverage, with an incremental cost-effectiveness ratio (ICER) of $421 per disability-adjusted life year (DALY) averted. 10/11 scenarios along the expansion path included community radiographic screening at progressively higher coverage, combined with a clinic-based approach. Combining complementary approaches to reach both groups at increased risk of TB (e.g. clinic-based screening) and groups with lower screening coverage (e.g. community-based screening) may increase cost-effectiveness of TB screening, compared to standalone approaches. When designing TB screening strategies, both population risk and existing screening coverage should be considered.

10.
arXiv (CS.CV) 2026-06-25

FlowID : Enhancing Forensic Identification with Latent Flow-Matching Models

Every day, many people die under violent circumstances, whether from crimes, war, migration, or climate disasters. Medico-legal and law enforcement institutions document many portraits of the deceased for evidence, but cannot immediately carry out identification on them. While traditional image editing tools can process these photos for public release, the workflow is lengthy and produces suboptimal results. In this work, we leverage advances in image generation models, which can now produce photorealistic human portraits, to introduce FlowID, an identity-preserving facial reconstruction method. Our approach combines single-image fine-tuning, which adapts the generative model to out-of-distribution injured faces, with attention-based masking that localizes edits to damaged regions while preserving identity-critical features. Together, these components enable the removal of artifacts from violent death while retaining sufficient identity information to support identification. To evaluate our method, we introduce InjuredFaces, a novel benchmark for identity-preserving facial reconstruction under severe facial damage. Beyond serving as an evaluation tool for this work, InjuredFaces provides a standardized resource for the community to study and compare methods addressing facial reconstruction in extreme conditions. Experimental results show that FlowID outperforms state-of-the-art open-source methods while maintaining low memory requirements, making it suitable for local deployment without compromising data privacy.

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

Breaking the Code: Security Assessment of AI Code Agents Through Systematic Jailbreaking Attacks

arXiv:2510.01359v2 Announce Type: replace-cross Abstract: Code-capable large language model (LLM) agents are embedded in software engineering workflows where they can read, write, and execute code, raising "jailbreak" stakes beyond text-only settings. Prior evaluations emphasize refusal or harmful-text detection, leaving open whether agents compile and run malicious programs. We present JAWS-Bench (Jailbreaks Across WorkSpaces), a benchmark spanning three escalating workspace regimes mirroring attacker capability: empty (JAWS-0), single-file (JAWS-1), and multi-file (JAWS-M). We pair this with a hierarchical, executable-aware Judge Framework that tests (i) compliance, (ii) attack success, (iii) syntactic correctness, and (iv) runtime executability, to measure deployable harm. Across seven LLM backends from five families, prompt-only attacks in JAWS-0 achieve 61% compliance; 58% are harmful, 52% parse, and 27% run end-to-end. In JAWS-1, compliance reaches ~100% for stronger models with a mean ASR (Attack Success Rate) ~71%; JAWS-M raises mean ASR to ~75%, with 32% runnable attack code. Wrapping an LLM in an agent increases ASR by 1.6$\times$, by overturning initial refusals during planning and tool use. Similar trends hold for OpenHands, SWE-Agent, and OpenAI Codex, suggesting our JAWS-Bench is agent-agnostic. Category analyses identify which attack classes are most vulnerable and deployable, motivating execution-aware defenses and refusal-preserving agent designs.

12.
PLOS Medicine 2026-06-02

Prognostic value of cervical length for spontaneous preterm birth in asymptomatic women with singleton pregnancy: An individual participant data meta-analysis

作者:

by Kelly Hughes, David Nguyen, Mason Aberoumand, Heather Ford, Erin Clarke, Nuria Banos Lopez, Margaret Dziadosz, Richard Fischer, Renato T. Souza, Jose Guilherme Cecatti, Kelly Orzechowski, Courtney Olson-Chen, Alberto Borges Peixoto, Vorapong Phupong, Joshua Rosenbloom, Moeun Son, Athena Souka, Liu Du, Michael Sean Esplin, Roberta Granese, Simi Gupta, Brenda Kazemier, Lindsay Kindinger, Pihla Kuusela, Jeanine Van der Ven, Omer Weitzner, Evelyn Minis, Alba Farras Llobet, Heather Frey, Rashmi Bagga, Siddhidatri Mishra, Elizabeth Patberg, Philip Bennett, Megan Hall, Andrew Shennan, Shaun Brennecke, Shakila Thangaratinam, Anna Lene Seidler, Ben Willem Mol, Rui Wang Background Spontaneous preterm birth (SPTB) is the leading cause of perinatal and early childhood mortality worldwide. Studies have generally suggested that mid-trimester transvaginal sonographic cervical length

13.
arXiv (CS.LG) 2026-06-18

Reliable Neural-Codec Text-to-Speech by ASR Self-Verification and Distillation: Near-Zero Catastrophic Failures Across Models and Codecs

arXiv:2606.18323v1 Announce Type: cross Abstract: Open autoregressive neural-codec text-to-speech (TTS) models sound excellent on typical inputs yet suffer stochastic catastrophic failures: on a meaningful fraction of utterances they emit silence, terminate early, or collapse into repetitive or hallucinated content. We show this failure mode is cheap to remove. Under a single format-robust metric (a catastrophic-failure rate via an ASR round-trip), best-of-N ASR self-verification drives failures to near-zero: no observed failures remain by N=2 on a standard corpus (LibriSpeech) and by N=4 on a hard prompt set. This is not an artifact of one model: the reduction replicates across four open codec-TTS systems and three neural codecs (XCodec2, SNAC, Mimi), reaching the near-zero floor by N=2 on three of the four. We then make the fix free at inference time by distilling the self-verified behaviour into the model, which recovers much of the robustness in single-shot decoding, closing ~52-58% of the failure mass on hard inputs at no test-time cost. The distillation gain concentrates where it is needed (hard inputs); on already-reliable prose there is no headroom and no detectable change. A controlled comparison adds a clean negative: offline direct preference optimization (DPO/IPO) does not beat plain supervised distillation, and an online iterative variant is promising but not statistically separable at our evaluation size. We report honestly the one model that resists (a larger Llasa where scale did not obviously help) and a rare-word capability ceiling that no self-distillation method overcomes

14.
arXiv (CS.CV) 2026-06-15

What Drives Test-Time Adaptation for CLIP? A Controlled Empirical Study from an Update Perspective

Vision-Language Models (VLMs) such as CLIP have become a standard backbone for open-vocabulary recognition, yet their zero-shot predictions remain vulnerable to distribution shifts encountered at deployment. Test-Time Adaptation (TTA) has recently been extended to CLIP as a lightweight solution, leading to a rapidly growing body of TTA4CLIP methods. However, empirical progress in this area has largely outpaced our understanding of what truly drives adaptation, where their gains originate, and under which shifts they remain reliable. In this paper, we take a step back from the pursuit of state-of-the-art accuracy and conduct a systematic controlled study of TTA4CLIP. We first organize existing methods into three unified paradigms according to what is updated at test time. We then introduce TTABC, an open-source TTA Benchmark for CLIP, which standardizes evaluation protocols and integrates more than 20 representative methods. Our controlled empirical analysis focuses on three key areas. First, we determine the driving factors in parameter-based methods, revealing that adaptation gains are primarily driven by test-time evidence and reliable proxies rather than heavy optimization. Second, we explore evidence utilization beyond heavy parameter tuning, showing that competitive and efficient performance can be achieved through cross- or current-sample evidence and lightweight prototype updates. Finally, we demonstrate that there is no silver bullet for TTA: no single adaptation paradigm is universally optimal, and the preferred paradigm depends on the nature of shift. We hope our benchmark and study provide a clearer understanding of the current TTA4CLIP landscape and establish a foundation for further research.

15.
arXiv (quant-ph) 2026-06-16

Quantum Measurement and Continuous Markov Processes

arXiv:2606.15958v1 Announce Type: new Abstract: These are the lecture notes for a course on diffusive quantum measuring instruments. They were prepared and delivered at the Perimeter Institute on Mondays and Thursdays, from 2:30 to 4:00 PM, beginning October 27th, 2025 and ending December 11th, 2025. These lectures were recorded and can be found at https://pirsa.org/c25038.

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

SLU-2K: A Question-Based Benchmark for Semantic Evaluation of Sign Language Translation

Sign Language Translation (SLT) is typically evaluated with surface-form metrics such as BLEU and ROUGE, which reward lexical overlap but do not directly measure whether a translation preserves the meaning of the source sign sequence. This is in contrast with the final objective of integrating SLT in assistive technology. In this work, we shift the focus from Sign Language Translation (SLT) to Sign Language Understanding (SLU), with particular emphasis on semantic understanding. Specifically, we evaluate systems based on their ability to correctly recover, from the input video, key semantic aspects of the original sentence, such as actions taking place and facts about people and objects. To enable this evaluation systematically, we propose SLU-2K, a dataset of 2,350 closed-ended video question-answer pairs based on the popular PHOENIX-2014T and CSL-Daily datasets. To obtain SLU-2K, we propose and extensively evaluate an automated data generation pipeline which produces questions across 7 categories, namely actions, locations, numbers, objects, people, time, and weather conditions. We show the potential of SLU-2K by evaluating popular Multimodal Large Language Models (MLLMs) and two representative state-of-the-art systems, MMSTL and SpaMo. Our results show that MLLMs reach near-random performance, highlighting the need for a more systematic integration of SLU in current AI systems. Furthermore, state-of-the-art translation systems carefully fine-tuned on in-domain data still exhibit a substantial semantic gap, with results ranging from 56.7% to 75.2%. These findings suggest that current SLT evaluation protocols overestimate true understanding and that future progress should be measured not only by fluency and n-gram overlap, but also by semantic correctness. Code, prompts, and benchmark files are available at https://github.com/ZenoTsT/SLU-2K

17.
medRxiv (Medicine) 2026-06-19

Within-host pathogen population diversity predicts treatment response in tuberculosis

Background: Tuberculosis (TB) treatment outcomes remain suboptimal, and standard clinical diagnostics cannot reliably identify patients at high risk of treatment failure or relapse at the time of diagnosis. While within-host Mycobacterium tuberculosis genetic diversity is hypothesized to reflect the viable bacterial burden and adaptive capacity of the infection, its clinical prognostic value remains unknown. Methods: We conducted a prospective cohort study of 364 patients with newly diagnosed, rifampicin-susceptible pulmonary TB in South Africa. Patients received standard 6-month therapy and were monitored for up to two years to ascertain composite unfavorable outcomes (treatment failure, death, or relapse). To accurately detect low-frequency (unfixed) genetic variants and eliminate reference bias artifacts, we mapped medium to high depth short-read sequences against matched, patient-specific long-read assemblies. The association between baseline pathogen genetic diversity and clinical outcomes was evaluated using multivariable Cox proportional-hazards models. Results: After bioinformatic filtering, true unfixed variants were relatively rare but significantly enriched in genes mediating pathogen adaptation and drug tolerance, including transporter proteins and two-component regulatory systems. Within-host bacterial genetic diversity (i.e., the total number of unfixed variants) ranged from 0-20, with a median of 1 per patient. In survival analysis adjusting for known clinical risk factors–including HIV status, prior TB, baseline smear positivity, and radiographic lung involvement–baseline within-host genetic diversity emerged as a strong, independent predictor of unfavorable treatment outcomes. For patients with greater than 3 unfixed variants at diagnosis, each increase of 5 unfixed variants was associated with more than double the risk of a composite unfavorable outcome (adjusted Hazard Ratio, 2.36; 95% CI, 1.27 to 4.39; p=0.007). Conclusions: Baseline within-host pathogen genetic diversity is an independent predictor of unfavorable TB treatment outcomes. As sequencing becomes increasingly integrated into routine diagnostics, quantifying unfixed variants is an accessible approach that promises to risk-stratify patients and guide the duration of individualized regimens.

18.
arXiv (CS.CL) 2026-06-17

The Critical Role of Model Selection in Causal Inference: A Comparative Analysis of Classification Models within the InferBERT Framework for Pharmacovigilance

Distinguishing causal adverse drug events (ADEs) from spurious correlations remains a central challenge in pharmacovigilance. The InferBERT framework integrates transformer models with Do-calculus, but its success hinges on the underlying classification model. This study evaluates the impact of model choice in InferBERT, assessing whether simpler models suffice, if domain-specific pre-training helps, whether scaling to LLMs improves causal detection, and the effect of post-hoc calibration. We performed a comparative study on two benchmarks: Analgesics-induced Acute Liver Failure (AILF) and Tramadol-related Mortalities (TRAM). Four models were evaluated-XGBoost (baseline), ALBERT (original InferBERT), BioBERT (biomedical transformer), and Med-LLaMA (medical LLM)-using 5-fold cross-validation repeated over 20 runs. We measured accuracy, Expected Calibration Error (ECE) pre- and post-isotonic regression, and Jaccard concordance of causal terms with PRR, ROR, and EBGM; significance was tested with paired t-tests. BioBERT achieved the highest accuracy on both datasets, while Med-LLaMA underperformed despite its size and parameter-efficient fine-tuning. Domain-specific pre-training was decisive. Calibration improved ECE but had mixed effects on accuracy and causal discovery. BioBERT's superiority also yielded the strongest concordance with traditional pharmacovigilance signals. These results show that domain-specific pre-training provides a clear advantage over simpler baselines and larger LLMs. Investing in manageable, domain-aware models is more effective for computational pharmacovigilance than simply scaling model size.

19.
PLOS Computational Biology 2026-06-18

Ten simple rules for turning your qualifying exam into an NIH-style fellowship proposal: A guide for graduate students

by Courtney Peña-Lima, Cameron S. Bader, Brendan K. Ball, Troy C. Dildine, Mekhala V. Dissanayake, Iris van ‘t Erve, Albina Ibrayeva, Amy Nippert, M.K. Quinn, Chelse Spinner, Samuel Thompson, Antonio Tomasso, Crystal M. Botham Qualifying exams, often referred to as “quals” or candidacy exams, are an important milestone in doctoral programs. Although the style of quals varies greatly by program and institution, it is usually a proposal that requires students to develop research ideas as well as their scientific writing skills. Many quals are modeled after funding mechanisms that graduate students can apply to and on a topic that the student will pursue in their dissertation. This paper offers graduate students a step-by-step guide on how to turn their quals into a fellowship-style research proposal, using National Institutes of Health (NIH) mechanisms as a benchmark, as this is the norm within US research institutions. This paper will be most useful for students who have completed or are in the process of completing proposal-based qualifying exams, usually in the second year of a doctoral program.

20.
arXiv (CS.AI) 2026-06-16

Rescaling Confidence: What Scale Design Reveals About LLM Metacognition

arXiv:2603.09309v2 Announce Type: replace Abstract: Verbalized confidence, in which LLMs report a numerical certainty score, is widely used to estimate uncertainty in black-box settings, yet the confidence scale itself (typically 0–100) is rarely examined. We show that this design choice is not neutral. Across six LLMs and three datasets, verbalized confidence is heavily discretized, with more than 78\% of responses concentrating on just three round-number values. To investigate this phenomenon, we systematically manipulate confidence scales along three dimensions: granularity, boundary placement, and range regularity, and evaluate metacognitive sensitivity using $meta-d'$. We find that a 0–20 scale consistently improves metacognitive efficiency over the standard 0–100 format, while boundary compression degrades performance and round-number preferences persist even under irregular ranges. These results demonstrate that confidence scale design directly affects the quality of verbalized uncertainty and should be treated as a first-class experimental variable in LLM evaluation.

21.
arXiv (CS.CV) 2026-06-24

DiffusionBench: On Holistic Evaluation of Diffusion Transformers

Diffusion transformer (DiT) research on image generation has converged to a single evaluation setup: class-conditional generation on ImageNet. While methods improve the FID and related metrics, it is increasingly unclear whether they reflect real progress in generative modeling. The natural alternative, i.e., text-to-image (T2I) generation, is perceived as too costly or inconvenient to train and evaluate and is often skipped. We argue that this perception no longer holds. We introduce NanoGen, a unified DiT training and evaluation framework. NanoGen matches state-of-the-art DiT baselines on ImageNet and, with 12 lines of configuration change, also trains competitive text-to-image models. It currently supports RAE, VAE, pixel-space, and MeanFlow diffusion methods under both ImageNet and T2I setups. Under NanoGen, training T2I requires comparable compute to ImageNet. After training 21 latent diffusion models with NanoGen, we observe that method ranking shows no strong correlation between ImageNet and T2I generation: Pearson correlation is between -0.377 and -0.580 across three metrics. This suggests that a method which improves class-conditional ImageNet FID may show no corresponding improvement on T2I, clearly indicating the necessity of evaluating DiTs on both tasks. To this end, we summarize ImageNet and text-to-image results, which yields DiffusionBench, a holistic benchmark for DiT research. We recommend reporting DiffusionBench in place of ImageNet alone: methods that improve DiffusionBench are more likely to reflect broader progress.

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

Breaking the Solver Bottleneck: Training Task Generators at the Learnable Frontier

The limiting resource for training agents via reinforcement learning (RL) is increasingly frontier task supply: valid, solvable tasks just difficult enough to train the current model. As reasoning and agentic models improve, fixed task distributions saturate, while naive synthetic generation yields tasks that are trivial, impossible, or ill-posed. Training a task generator with RL to optimize validity and learnability can address this bottleneck, but direct optimization requires repeated solver rollouts per candidate. For software-engineering (SWE) tasks, a single rollout can take tens of minutes; solver-in-the-loop generator training is intractable. We introduce PROPEL, a solver-amortized framework for training task generators at the targeted solve rate. PROPEL trains a lightweight activation probe on a one-time labeled corpus of generated tasks and solver outcomes. The probe predicts target-solver pass rate from a frozen generator reference model and serves as a proxy for solve rate during generator optimization, reducing generator evaluation to a single forward pass. Across math, code, and software-engineering at multiple model scales, PROPEL shifts generation toward the targeted solve rate: for coding, tasks generated at the learnable frontier increase from $10.1\% \rightarrow 20.0\%$ for a Qwen2.5-3B-Instruct solver and from $5.3\% \rightarrow 12.6\%$ for a Qwen2.5-7B-Instruct solver. For SWE, PROPEL increases the share of generations at the targeted solve rate from $9.8\% \rightarrow 19.6\%$ for Qwen3.5-27B on repositories not seen during training of probe and generator.

23.
arXiv (CS.LG) 2026-06-11

Mitigating Disparate Impact of Differentially Private Learning through Bounded Adaptive Clipping

arXiv:2506.01396v2 Announce Type: replace Abstract: Differential privacy (DP) has become an essential framework for privacy-preserving machine learning. Existing DP learning methods, however, often have disparate impacts on model predictions, e.g., for minority groups. Gradient clipping, which is often used in DP learning, can suppress larger gradients from challenging samples. We show that this problem is amplified by adaptive clipping, which will often shrink the clipping bound to tiny values to match a well-fitting majority, while significantly reducing the accuracy for others. We propose bounded adaptive clipping, which introduces a tunable lower bound to prevent excessive gradient suppression. Our method improves worst-class accuracy by over 10 percentage points on Skewed and Fashion MNIST compared to unbounded adaptive clipping, 7 points compared to Automatic clipping, and 5 points compared to constant clipping. The code is available at https://github.com/TrustworthyMLHelsinki/adaptive-clipping-fairness.

24.
arXiv (CS.LG) 2026-06-16

Next-Latent Prediction Transformers Learn Compact World Models

arXiv:2511.05963v4 Announce Type: replace Abstract: Transformers replace recurrence with a memory that grows with sequence length and self-attention that enables ad-hoc lookups over past tokens. Consequently, they lack an inherent incentive to compress history into compact latent states with consistent transition rules. This often leads to learning solutions that generalize poorly. We introduce Next-Latent Prediction (NextLat), which extends standard next-token training with self-supervised predictions in the latent space. Specifically, NextLat trains a transformer to learn latent representations that are predictive of its next latent state given the next token. Theoretically, we show that these latents provably converge towards belief states, compressed information about the history necessary to predict the future. This simple auxiliary objective injects a recurrent inductive bias into transformers while leaving their architecture, parallel training efficiency, and inference unchanged. NextLat effectively encourages transformers to form compact internal world models with coherent belief states and transition dynamics – crucial properties not guaranteed by standard next-token prediction alone. Empirically, across benchmarks in world modeling, reasoning, planning, and language modeling, NextLat demonstrates significant gains over standard next-token prediction and other baselines in downstream accuracy, representation compression, and lookahead planning. Furthermore, NextLat enables variable-length self-speculative decoding, accelerating inference by up to 3.3x in language modeling. NextLat offers a simple yet effective paradigm for learning compact, predictive representations in transformers that generalize better. Our code is available at https://github.com/JaydenTeoh/NextLat.

25.
arXiv (CS.CL) 2026-06-15

Beyond Rubrics: Exploration-Guided Evaluation Skills for Reward Modeling

Open-ended reward modeling requires judges that can follow subtle, domain-specific preferences when verifiable answers are unavailable. Existing rubric-based methods often address this by generating criteria online for each query, but the extra generation step can add inference overhead and produce rigid or misaligned guidance. We introduce Eval-Skill, an exploration-guided method that synthesizes reusable evaluation skills for reward modeling and reframes reward guidance as context evolution rather than parameter training or per-query rubric generation. Using only 100 cases per domain for skill evolution, Eval-Skill synthesizes reusable domain-level evaluation skills through two progressive stages, workflow generation followed by principle generation, with exploration and selection interleaved across both stages. Once generated, a skill is directly injected into the judge context. Across multiple RM benchmarks, Eval-Skill consistently improves diverse judge backbones; on RewardBench 2, it yields significant gains over vanilla judging for each main backbone (+13.44% for Qwen3-8B, and 18.51% for DeepSeek-V4-Flash). Further analyses of evolution-time scaling, generalizability, and transferability show that compact evaluation skills offer an efficient new paradigm for LLM-based evaluation. Code is available at https://github.com/xing-stellus-yue/Eval-Skill.