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) 2026-06-11

Incremental costs of transitioning from four to eight WHO-recommended antenatal care visits in Uganda: A costing analysis from a societal perspective

Background In 2016, the World Health Organization revised its antenatal care (ANC) recommendation from four to eight visits. For low- and middle-income countries like Uganda, where achieving even four visits remains a challenge, this transition has significant cost implications for both the health system and households. This study estimated the incremental costs of adopting the eight-visit model from a societal perspective. Methods The study was conducted in six government health facilities in southwestern Uganda. A micro-costing approach estimated health facility costs (personnel, equipment, consumables, and overhead). Costs incurred at patients end (transport, ultrasound, medical expenses, and time) were collected from 785 women using a questionnaire, with all costs in 2025 USD. Results For an average of 4.3 visits, total cost per woman was $100.1: facility costs $43.7 (43.7%), and patient costs $56.4 (56.3%). Transitioning to eight visits would increase total cost by $57.8 (57.8%), of which $36.4 (63.0%) would fall on households, equivalent to 68.8% of average monthly household income. Total costs would rise by 55.4% ($115.5 to $179.5) at Health Center IVs and 64.3% ($102.3 to $168.1) at Health Center IIIs, with facility costs up 43.4% and 62.9% and patient costs up 61.2% and 65.7%, respectively. Conclusion Transitioning to eight ANC visits would impose a large financial burden on households, with the incremental patient cost equivalent to more than two-thirds of average monthly household income. Equitable implementation requires improving availability of medicines and diagnostics, subsidizing transport, exploring telemedicine or community-based models, and improving efficiency at lower-tier health centers.

02.
arXiv (CS.CL) 2026-06-18

MemRerank: Preference Memory for Personalized Product Reranking

LLM-based shopping agents increasingly rely on long purchase histories and multi-turn interactions for personalization, yet naively appending raw history to prompts is often ineffective due to noise, length, and relevance mismatch. We propose MemRerank, a preference memory framework that distills user purchase history into concise, query-independent signals for personalized product reranking. To study this problem, we build an end-to-end benchmark and evaluation framework centered on an LLM-based 1-in-5 selection task, which measures both memory quality and downstream reranking utility. We further train the memory extractor with reinforcement learning (RL), using downstream reranking performance as supervision. Experiments with two LLM-based rerankers show that MemRerank consistently outperforms no-memory, raw-history, and off-the-shelf memory baselines, yielding up to +10.61 absolute points in 1-in-5 accuracy. These results suggest that explicit preference memory is a practical and effective building block for personalization in agentic e-commerce systems.

03.
arXiv (CS.CV) 2026-06-18

Zero-Shot Cross-City Generalization in End-to-End Autonomous Driving: Self-Supervised versus Supervised Representations

End-to-end autonomous driving models are typically trained on multi-city datasets using supervised ImageNet-pretrained backbones, yet their ability to generalize to unseen cities remains largely unexamined. When training and evaluation data are geographically mixed, models may implicitly rely on city-specific cues, masking failure modes that would occur under real-world domain shifts when generalizing to new locations. In this work, we formulate zero-shot cross-city transfer as a controlled representation-level stress test for end-to-end autonomous driving and ask how visual pretraining affects transfer behavior under geographic domain shift. We conduct a comprehensive study by integrating self-supervised backbones I-JEPA, DINOv2, and MAE into planning frameworks. We evaluate performance under strict geographic splits on nuScenes in the open-loop setting and on NAVSIM in the closed-loop evaluation protocol. Our experiments reveal a substantial generalization gap when transferring models across cities with different road topologies, traffic conventions, and visual environments. In open-loop evaluation, a supervised backbone exhibits severe degradation when transferring between cities, yet some domain-specific self-supervised methods can substantially reduce both displacement and collision degradation. In closed-loop evaluation, self-supervised pretraining improves average out-of-distribution PDMS in several single-city training settings. Our results provide empirical evidence that representation learning influences the robustness of cross-city planning and motivate zero-shot geographic transfer as an important stress test for evaluating end-to-end autonomous driving systems.

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

Petrov-Galerkin Variational Physics-Informed Neural Network Framework for Two-Dimensional Singularly Perturbed Problems

arXiv:2606.16510v1 Announce Type: cross Abstract: This study proposes a Petrov-Galerkin based Variational Physics-Informed Neural Network (VPINN) for efficiently solving two-dimensional singularly perturbed problems (SPPs) with one and two small perturbation parameters. The approach employs neural networks to construct the trial solution space, while tensor-product hat functions are adopted as test functions to enforce the variational form. To accurately resolve of sharp boundary layers, the variational form is implemented using a Petrov-Galerkin formulation. Dirichlet boundary conditions are imposed directly, while the source terms are computed using automatic differentiation. Computational experiments on standard two-dimensional problems demonstrate that the proposed method achieves high accuracy in both the maximum and L_2 norms. These results confirm the efficiency and robustness of the Petrov-Galerkin VPINN approach in accurately capturing the multiscale features of two-dimensional SPPs.

05.
bioRxiv (Bioinfo) 2026-06-11

inquiSTR: a toolkit for accurate and efficient population-scale tandem repeat genotyping and analysis

Tandem repeats are highly mutable genomic elements linked to human traits and diseases. Profiling large catalogs of tandem repeats from population-scale long-read sequencing data requires accurate and efficient tools. We introduce inquiSTR, a command-line toolkit for fast genome-wide tandem repeat length genotyping. inquiSTR, with efficient parallel processing and low-memory streaming algorithms, genotypes a genome-wide repeat catalog of 1.78 million loci in less than two minutes. Benchmarking shows high accuracy and significantly faster performance compared to existing tools and truth sets. inquiSTR also provides methods for downstream analyses such as population structure inference, association testing, and outlier detection.

06.
medRxiv (Medicine) 2026-06-15

Validating Field-Feasible Measures of Recent Khat Use: A Diagnostic Accuracy Study Comparing Amphetamine Immunoassay and Assisted Self-Report Against HPLC in an Ethiopian Male Cohort

Background: Khat (Catha edulis) is a widely consumed natural amphetamine-analog used across East Africa and the Arabian Peninsula. Accurate field-feasible measurement of recent khat use is a prerequisite for large-scale epidemiological research; yet no validated alternatives to laboratory reference methods have been identified in the scientific literature. This nested validation study evaluated the diagnostic accuracy of two point-of-care measures, a commercial amphetamine immunoassay and a Timeline Followback (TLFB) Assisted Self-Report (ASR), against high-performance liquid chromatography (HPLC) quantification of urinary norephedrine (NE), while additionally assessing agreement between the two field measures. Methods: A prospective, random sub-sample of 119 male participants aged 18-40 years from the Gilgel Gibe Field Research Center (GGFRC) longitudinal cohort, Ethiopia (validation timepoint T2, 2015), was used. Three index-reference comparisons were conducted: (1) amphetamine immunoassay (nal von minden, Drug-Screen AMP test, 300 ng/mL cutoff) vs. HPLC; (2) binary ASR (past-week use) vs. HPLC; and (3) binary ASR vs. immunoassay. Sensitivity (positive percent agreement, PPA), specificity (negative percent agreement, NPA), positive predictive value (PPV), negative predictive value (NPV), overall accuracy (overall percent agreement, OPA), and Cohen's kappa were calculated with 95% confidence intervals. Pre-specified secondary analyses applied three pharmacokinetically-informed recall windows (0-2, 3-5, and 6-7 days prior to interview) to ASR. Results: Against HPLC (77 positive, 42 negative), the immunoassay showed perfect specificity (1.0 [0.916-1.0]) and PPV (1.0 [0.91-1.0]) but low sensitivity (0.52 [0.40-0.64]), NPV (0.53 [0.42-0.65]), overall accuracy (0.69 [0.60-0.77]), and weak kappa (0.43 [0.34-0.52]). Binary ASR showed high sensitivity (0.96 [0.89-0.99]), specificity of 0.60 [0.433-0.74], PPV (0.81 [0.72-0.89]), NPV (0.89 [0.72-0.98]), with overall accuracy 0.83 [0.75-0.89] and moderate kappa (0.60 [0.51,0.69]). Restricting ASR to use within 0-2 days improved specificity to 0.69 [0.52-0.84], PPV to 0.86 [0.77-0.93], overall accuracy to 0.87 [0.79-0.93], and kappa to 0.69 [0.61-0.78] (moderate), while sensitivity (0.96 [0.89-0.99]) and NPV (0.89 [0.72-0.98]) remained stable. Against the immunoassay, ASR achieved high PPA of (1.0 [0.91-1.0]), NPA of 0.35 [0.25-0.47], OPA of 0.57 [0.48-0.66], and minimal kappa (0.27 [0.19-0.35]). Conclusions: Time-stratified ASR (0-2 days) is a valid, scalable alternative to biological testing for recent khat use in resource-limited settings. The immunoassay's 300 ng/mL cutoff functions as a marker of heavy or recent high-dose khat use rather than any-use detection. Its perfect specificity and PPV make it valuable as a confirmatory test for substantial exposure, while its lower sensitivity reflects calibration to amphetamine rather than to khat-derived cathinone metabolite. Keywords: khat; Catha edulis; diagnostic accuracy; STARD; self-report; immunoassay; HPLC; Ethiopia; substance use measurement

07.
arXiv (CS.CL) 2026-06-15

AgentSpec: Understanding Embodied Agent Scaffolds Through Controlled Composition

LLM agents are increasingly built not as single model calls, but as scaffolded systems that combine reasoning, memory, reflection, action execution, and learning. While such scaffolds often improve performance, they are often embedded in tightly coupled pipelines, making it difficult to isolate component contributions, compare alternative designs, or understand how module interactions shape agent behavior. We introduce AgentSpec, a modular specification framework that represents embodied agents as typed compositions of reusable policy components with standardized interfaces. AgentSpec standardizes the interfaces among perception, memory, reasoning, reflection, action, and optional learning, enabling components to be swapped and recombined under controlled conditions. We instantiate this framework across DeliveryBench, ALFRED, MiniGrid, and RoboTHOR, and analyze reasoning, memory, reflection, and reinforcement-learning modules across model backbones. Our results show that agent performance is governed by scaffold compatibility and interaction effects rather than isolated module strength. In particular, structured multi-granularity memory improves long-horizon state tracking, reasoning and memory interact non-uniformly across environments, reflection trades off correction and cost, and RL-trained policies compose best when optimized with deployment-time scaffold structure. AgentSpec provides a controlled foundation for studying, comparing, and designing composable LLM agents. Our code, baselines and interactive playground are publicly available at https://agentspec-embodied.github.io.

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

How to Score Experts for One-Shot MoE Expert Pruning: A Unified Formulation and Selection Principle

arXiv:2606.15716v1 Announce Type: new Abstract: Mixture-of-Experts (MoE) language models reduce per-token computation through sparse expert activation, yet deployment still requires storing the full expert pool, making one-shot expert pruning a practical approach for reducing memory usage. Although effective, existing criteria are largely heuristic, and no single criterion is universally optimal. Thus, establishing a principle for selecting pruning criteria suited to different deployment objectives remains an important yet largely underexplored problem in one-shot expert pruning. To this end, we introduce a unified formulation for one-shot MoE expert pruning organized around three factors: routing frequency, gate weighting, and activation strength. The formulation yields a criteria selection principle: task-agnostic pruning should favor routed-token-averaged, gate-free activation-based criteria, whereas task-specific pruning can benefit from retaining routing-frequency and gate-weight information. Beyond this principle, the formulation also provides a systematic view of existing heuristic criteria and gives rise to two new task-agnostic criteria, Mean Activation Norm (MAN) and Mean Squared Activation Norm (MSAN). Across four representative MoE models and 16 diverse benchmarks, MAN and MSAN are consistently strong in the task-agnostic setting, obtain the top-two average ranks, and improve average performance by up to 8.8 points over the strongest baseline.

09.
arXiv (CS.AI) 2026-06-12

PhononBench:A Large-Scale Phonon-Based Benchmark for Dynamical Stability in Crystal Generation

arXiv:2512.21227v3 Announce Type: replace-cross Abstract: In recent years, generative artificial intelligence has made significant advances in the design of crystalline materials, giving rise to approaches based on graph neural networks, diffusion models, and large language models. Existing evaluations commonly follow the stability-uniqueness-novelty (S.U.N.) framework, where stability is primarily assessed using thermodynamic criteria, which do not fully capture the dynamical stability essential for a material's practical existence. Dynamical stability is a key determinant of whether a material can be synthesized and persist, with phonon spectrum calculations serving as the standard for its evaluation. However, the high computational cost of such calculations has prevented large-scale assessment of dynamical stability in generated crystals. In this work, we introduce PhononBench, the first large-scale benchmark for dynamical stability in AI-generated crystals. Leveraging the recently developed MatterSim interatomic potential, which achieves density-functional-theory (DFT)-level accuracy in phonon predictions across more than 10,000 materials, PhononBench enables efficient phonon calculations and dynamical-stability analysis for 133,838 crystal structures generated by 7 leading crystal generation models. PhononBench reveals a widespread limitation of current generative models: unless otherwise specified, all reported dynamical-stability metrics are evaluated at a phonon-frequency threshold of -0.1 THz, with the average dynamical-stability rate across all generated structures being only 32.15%, and the top-performing model, MatterGen, reaching just 45.05%.In addition, we identify 32,995 crystal structures that are phonon-stable across the entire Brillouin zone under a strict threshold of -0.001 THz. In addition, a web-based service is accessible at http://phononbench.cn/, enabling minute-level ultra-fast phonon predictions.

10.
arXiv (math.PR) 2026-06-15

Stationary measures for higher spin vertex models on a strip

Authors:

arXiv:2309.04897v2 Announce Type: replace-cross Abstract: We introduce a higher spin vertex model on a strip with fused vertex weights. This model can be regarded as a generalization of both the unfused six-vertex model on a strip arXiv:2212.09111 and an 'integrable two-step Floquet dynamics' model introduced in arXiv:1711.08884. We solve for the stationary measure using a fused version of the matrix product ansatz and then characterize it in terms of the Askey-Wilson process. Using this characterization, we obtain the limits of the mean density along an arbitrary down-right path. It turns out that all these models share a common phase diagram, which, after an appropriate mapping, matches the phase diagram of open ASEP. This provides evidence for the universality of this phase diagram.

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

Plan, Don't Pose: Long Composite Motion Generation with Text-Aligned BFM

arXiv:2605.29906v2 Announce Type: replace Abstract: Text-to-motion (T2M) generation has broad applications in character animation, virtual avatars, and human-robot interaction. Existing methods typically generate pose trajectories or motion tokens directly from language, forcing a single model to handle semantic interpretation, long-horizon structure, and low-level physical realization. This coupling makes them costly and often unreliable for long, compositional, or semantically dense prompts. We propose Text2BFM, the first framework that aligns natural language with pretrained Behavioral Foundation Models (BFMs) for T2M generation without relying on heavy end-to-end motion generators. Text2BFM operates in the latent policy space of a frozen BFM, using it as an executable motion prior. A text-aligned variational behavioral bottleneck compresses BFM policy-latent sequences into compact motion representations that are compatible with language and preserve long-horizon behavioral structure. Generation is performed in this compact behavioral manifold with a lightweight conditional generator, and the resulting latent encoded behaviors are decoded into policy latents that drive the pretrained frozen BFM. By decoupling semantic planning from motion execution, Text2BFM achieves efficient, robust T2M generation and strong performance on long, compositional textual descriptions.

12.
arXiv (CS.LG) 2026-06-16

Dynestyx: A Probabilistic Programming Library for Dynamical Systems

arXiv:2606.16985v1 Announce Type: cross Abstract: State-space models (SSMs) are the standard formalism for Bayesian treatment of dynamical systems, with natural applications in statistics, signal processing, and machine learning. Despite their importance in both theory and application, dynamical systems have proven difficult to incorporate in modern probabilistic programming languages (PPLs), making state-of-the-art methods less accessible to practitioners and introducing friction in following the "Bayesian workflow." We introduce dynestyx, a probabilistic programming library with first-class support for SSMs, including state-of-the-art methods in the estimation of both states and parameters. Through a single, unified interface, users may specify arbitrary priors for discrete-time or continuous-time dynamical systems, perform inference over mixed-effect data, and make state and parameter estimates with principled uncertainty quantification.

13.
medRxiv (Medicine) 2026-06-11

Advancing Clinical Implementation of Cardiovascular Polygenic Risk Scores Through Patient-Level Robustness Assessment

Background and Aims: Polygenic risk scores (PRSs) for atherosclerotic cardiovascular disease (ASCVD) can perform equivalently at the population level yet disagree for individual patients. We examined whether such intra-individual variability reflects genuinely complementary risk information or mainly statistical and methodological uncertainty, and whether it affects clinical classification once PRSs are integrated into SCORE2-OP. Methods: In 4,137 ASCVD-free participants of the CoLaus|PsyCoLaus cohort (478 incident events over a median 14.4 years), we identified 16 ASCVD-PRSs with practically equivalent population-level performance using Bayesian equivalence testing. We quantified intra-individual variability (standard deviation, coefficient of variation, intraclass correlation, Cohen's kappa, extreme discordance), tested whether discordance exceeded chance, decomposed scores into shared and unique genetic components, and assessed variability after integration into SCORE2-OP, benchmarked against perturbation of systolic blood pressure. Results: For a typical individual, risk estimates varied by 18 percentile points across PRSs. Discordance matched chance expectations under a shared-signal model, with no distinct phenotypic profile among discordant individuals, and predictive power resided overwhelmingly in the shared genetic component. Variability tracked PRS size and weighting rather than distinct variants. After integration into SCORE2-OP, 75.6% of participants were placed in different categories by at least one model and 54.6% as both low and high risk; instability was concentrated near guideline thresholds and far exceeded that from blood-pressure measurement error. Conclusions: Equivalent population-level performance is not sufficient to treat PRSs as interchangeable at the individual level, and methodological standardisation and pragmatic clinical trials remain necessary to determine whether PRS integration improves long-term cardiovascular outcomes.

14.
arXiv (CS.LG) 2026-06-16

CADO: From Imitation to Cost Minimization for Heatmap-based Solvers in Combinatorial Optimization

arXiv:2602.08210v2 Announce Type: replace Abstract: Heatmap-based solvers have emerged as a promising paradigm for Combinatorial Optimization (CO). However, we argue that the dominant Supervised Learning (SL) training paradigm suffers from a fundamental objective mismatch: minimizing imitation loss (e.g., cross-entropy) does not guarantee solution cost minimization. We dissect this mismatch into two deficiencies: Decoder-Blindness (being oblivious to the non-differentiable decoding process) and Cost-Blindness (prioritizing structural imitation over solution quality). We empirically demonstrate that these intrinsic flaws impose a hard performance ceiling. To overcome this limitation, we propose CADO (Cost-Aware Diffusion models for Optimization), a streamlined Reinforcement Learning fine-tuning framework that formulates the diffusion denoising process as an MDP to directly optimize the post-decoded solution cost. We introduce Label-Centered Reward, which repurposes ground-truth labels as unbiased baselines rather than imitation targets, and Hybrid Fine-Tuning for parameter-efficient adaptation. CADO achieves state-of-the-art performance across diverse benchmarks, validating that objective alignment is essential for unlocking the full potential of heatmap-based solvers.

15.
arXiv (CS.AI) 2026-06-17

The Discrete-Log Clock: How a Transformer Learns Modular Multiplication

arXiv:2606.17399v1 Announce Type: cross Abstract: When small transformers grok modular multiplication, prior work reports that the learned embedding has a "dense" Fourier spectrum requiring all frequencies. This contrasts with modular addition, where only a sparse set of key frequencies suffices. We show this density is an artifact of analyzing in the wrong basis. The natural Fourier transform for multiplication is not the standard additive DFT but the multiplicative character transform, which decomposes functions on the multiplicative group $(\mathbb{Z}/p\mathbb{Z})^*$ into its irreducible representations. Applying this transform to a grokked transformer trained on $a \cdot b \bmod 113$, we find the embedding spectrum becomes highly sparse (Gini coefficient 0.58 vs. 0.07 in the additive basis) with only 4 key frequencies carrying significant energy. Furthermore, 96.9% of MLP neurons are cleanly tuned to a single multiplicative frequency, and neuron activation heatmaps reveal 2D-periodic structure when reordered by the discrete logarithm. These results demonstrate the transformer reduces multiplication to addition in discrete-log space, implementing a "Discrete-Log Clock" algorithm analogous to Nanda et al.'s Clock algorithm for addition. The methodology generalizes: matching the analysis basis to the algebraic structure of the task reveals interpretable structure where standard tools see noise.

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

Learning to Share: Selective Memory for Efficient Parallel Agentic Systems

arXiv:2602.05965v2 Announce Type: replace-cross Abstract: Agentic systems solve complex tasks by coordinating multiple agents that iteratively reason, invoke tools, and exchange intermediate results. To improve robustness and solution quality, recent approaches deploy multiple agent teams running in parallel to explore diverse reasoning trajectories. However, parallel execution comes at a significant computational cost: when different teams independently reason about similar sub-problems or execute analogous steps, they repeatedly perform substantial overlapping computation. To address these limitations, in this paper, we propose Learning to Share (LTS), a learned shared-memory mechanism for parallel agentic frameworks that enables selective cross-team information reuse while controlling context growth. LTS introduces a global memory bank accessible to all teams and a lightweight controller that decides whether intermediate agent steps should be added to memory or not. The controller is trained using stepwise reinforcement learning with usage-aware credit assignment, allowing it to identify information that is globally useful across parallel executions. Experiments on the AssistantBench and GAIA benchmarks show that LTS significantly reduces overall runtime while matching or improving task performance compared to memory-free parallel baselines, demonstrating that learned memory admission is an effective strategy for improving the efficiency of parallel agentic systems. Project page: https://joefioresi718.github.io/LTS_webpage/

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

Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention

arXiv:2510.04212v4 Announce Type: replace-cross Abstract: The pursuit of computational efficiency has driven the adoption of low-precision formats for training transformer models. However, this progress is often hindered by notorious training instabilities. This paper provides the first mechanistic explanation for a long-standing and unresolved failure case where training with flash attention in low-precision settings leads to catastrophic loss explosion. Our in-depth analysis reveals that the failure is not a random artifact but caused by two intertwined phenomena: the emergence of similar low-rank representations within the attention mechanism and the compounding effect of biased rounding errors inherent in low-precision arithmetic. We demonstrate how these factors create a vicious cycle of error accumulation that corrupts weight updates, ultimately derailing the training dynamics. To validate our findings, we introduce a minimal modification to the flash attention that mitigates the bias in rounding errors. This simple change stabilizes the training process, confirming our analysis and offering a practical solution to this persistent problem. Code is available at https://github.com/ucker/why-low-precision-training-fails.

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

SMART: A Flexible, Interpretable, and Scalable Spatio-temporal Brain Atlas from High-Resolution Imaging Data

We introduce SMART, a framework for learning a flexible, interpretable, and scalable spatio-temporal brain atlas from longitudinal high-resolution 3D medical images. Existing approaches to spatio-temporal atlas construction rely on black-box generative models that lack flexibility, limit interpretability, and struggle to scale to high-dimensional data. SMART addresses these challenges by learning a continuous disease-time atlas that decouples global group-wise disease dynamics from their patient-specific anatomical manifestation. Guided by anatomically inspired priors, SMART models interpretable global trajectories of regional progression along a shared disease timeline through region-specific differential equations. Global trajectories are further personalized to individual anatomies via dense diffeomorphic displacements parameterized by a flexible and scalable multi-scale Neural Cellular Automata. Evaluated on five longitudinal MRI datasets in Alzheimer's disease (ADNI-1/GO/2, OASIS-3, AIBL; > 1,300 subjects), SMART produces anatomically meaningful predictions of disease progression and achieves state-of-the-art forecasting accuracy and improved temporal consistency over adversarial and diffusion baselines. Our approach establishes a new paradigm for flexible, interpretable, and scalable modeling of spatio-temporal change in high-dimensional medical image time-series.

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

ReCal: Reward Calibration for RL-based LLM Routing

arXiv:2606.12479v1 Announce Type: cross Abstract: Large language model (LLM) routing has emerged as an effective paradigm for leveraging the complementary strengths of multiple LLMs through dynamic model and reasoning-strategy selection. Recent reinforcement learning (RL)-based routing methods further improve routing quality by optimizing routing policies from interaction feedback. However, they still struggle to provide informative and comparable learning signals under heterogeneous tasks with varying difficulty. In practice, multiple objectives (e.g., correctness, format behavior) are aggregated into a single scalar reward, leading to ambiguous credit assignment and conflicting optimization signals. Moreover, reward signals exhibit significant variability across instances, where some instances produce higher or more variable rewards, introducing optimization bias that favors trivial samples over informative ones. To address these issues, we propose ReCal, a \underline{Re}ward \underline{Cal}ibration framework for RL-based LLM routing. We first introduce a hierarchical reward decomposition mechanism with component-wise advantage estimation. We further propose a distribution-aware optimization strategy that calibrates optimization variability through variance-aware reweighting and per-dataset normalization. Experiments on seven datasets demonstrate that ReCal consistently improves routing performance, and training stability over baselines. Code is available at https://anonymous.4open.science/r/ReCal.

21.
arXiv (CS.CL) 2026-06-11

Reassessing High-Performing LLMs on Polish Medical Exams: True Competence or Bias-Driven Performance?

Large language models (LLMs) in medicine are mainly evaluated using multiple-choice question answering (MCQA), which can overestimate real clinical ability due to guessing strategies and answer biases. To address these limitations, we introduce an expanded and more challenging benchmark based on Polish medical exams, adding over 15,000 questions, two new domains, and four structural modifications that reduce MCQA-specific artifacts and better test reasoning. We evaluate 21 LLMs and show that evaluation design strongly affects results. Under our harder setup, the best model (Qwen3.5-122B) drops by 28.4 and 31 pp on English and Polish exams, respectively. Despite low evidence of data contamination, standard MCQA scores do not reliably reflect true medical competence. To facilitate further research, we make our benchmark publicly available.

22.
arXiv (quant-ph) 2026-06-17

Hamiltonian description of nonreciprocal interactions

arXiv:2505.05246v5 Announce Type: replace-cross Abstract: In a vast class of systems, which includes members as diverse as sedimenting particles and bird flocks, interactions do not stem from a potential, and are in general nonreciprocal. Thus, it is not possible to define a conventional energy function, nor to use analytical or numerical tools that rely on it. Here, we overcome these limitations by constructing a Hamiltonian that includes auxiliary degrees of freedom; when subject to a constraint, this Hamiltonian yields the original nonreciprocal dynamics. We show that Glauber dynamics based on the constrained Hamiltonian reproduce both stationary and nonstationary states of the original Langevin dynamics, as we explicitly illustrate for dissipative XY spins with vision-cone interactions. Further, the symplectic structure inherent to our construction enables us to apply the well-developed notions of Hamiltonian engineering, which we demonstrate by varying the amplitude of a periodic drive to tune the spin interactions between those of a square and a chain lattice geometry. Overall, our framework for generic nonreciprocal pairwise interactions paves the way for bringing to bear the full conceptual and methodological power of conventional statistical mechanics and Hamiltonian dynamics to nonreciprocal systems.

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

Demystifying Variance in Circuit Discovery of LLMs

arXiv:2606.16920v1 Announce Type: cross Abstract: Circuit discovery is a key technique in mechanistic interpretability to pinpoint the model components that are crucial for performing a given task. Although the current state-of-the-art method (EAP-IG) performs well on the metric of (un)faithfulness, it suffers from substantial variability. This includes resampling variance, where the circuit changes when we probe with a new batch of data from the same distribution; rephrasing variance, where the discovered circuit shifts when the prompts are rephrased; and sample-wise variance, where a circuit with low population unfaithfulness exhibits large fluctuations in unfaithfulness across individual samples. This paper studies the roots of these variances. We demonstrate that CEAP, our new circuit discovery method that improves upon EAP-IG with a theoretical guarantee, can substantially lessen resampling variance. We further show that rephrasing variance arises because prompts with different templates tend to activate different circuits in the model. This leads us to argue that it may be challenging to find a comprehensive circuit that explains and controls the model's behavior on a task, which can be expressed in countless templates, suggesting that LLMs may be inherently hard to steer. We show that sparsity, which has been claimed to form more compact and interpretable task circuits, fails to solve this problem. Regarding sample-wise variance, we argue that it is largely benign: extremely poor unfaithfulness scores often stem from how unfaithfulness is defined, rather than from defects in the measured circuits. We show that the magnitude of unfaithfulness is affected by selective contribution scaling, a neural mechanism that accounts for the extremely poor scores sometimes observed.

24.
arXiv (CS.LG) 2026-06-12

Generative Modeling of Bach-Style Symbolic Music: A Comparative Study of Autoregressive, Latent-Variable, and Adversarial Approaches

arXiv:2606.13626v1 Announce Type: cross Abstract: We study generative modeling of Bach-style symbolic piano music using a shared MIDI corpus and three model families: autoregressive LSTMs with attention, latent-variable models including recurrent VAEs and vector-quantized VAEs, and generative adversarial networks. We compare their ability to model polyphonic note sequences, learn useful latent representations, and generate stylistically coherent compositions. Our experiments show that the autoregressive LSTM with attention produces the most musically coherent samples, while vector quantization helps mitigate posterior collapse and yields more structured outputs than conventional recurrent VAEs. The adversarial approach captures local pitch patterns but remains difficult to train and generalizes less reliably to Bach's style. These results highlight the relative strengths and failure modes of autoregressive, latent-variable, and adversarial approaches for symbolic music generation.

25.
arXiv (CS.LG) 2026-06-17

CheckMIABench: Firm Foundations For Membership Inference Attacks on Language Models

arXiv:2606.17464v1 Announce Type: new Abstract: Membership inference attacks (MIAs) are a canonical way to assess a machine learning model's privacy properties. Although several attempts have been made to evaluate MIAs on language models, the extant literature has suffered numerous difficulties in constructing clean evaluations to test new techniques. In particular, subtle distribution shifts between member and non-member sets can undermine the statistical validity of MIAs; recent work has underscored this by showing that "blind" methods with no access to the underlying model can perform far better than published methods on the same benchmarks. This paper constructs a benchmark for principled evaluation of MIAs against LLMs, by leveraging the insight that training data before and after a fixed point during training are drawn from the same distribution. Therefore, all open-source models with intermediate checkpoints and public training data can be converted into MIA testbeds. We apply our framework to a half-dozen published attacks on the Pythia and OLMo family of models, from 70M to 7B parameters. To facilitate further privacy research, we open-source a modular library for designing and implementing attacks in this setting: https://github.com/safr-ai-lab/pandora_llm.