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) 2026-06-24

Food additive exposure associated with reduction in gut microbiota diversity

Consumption of ultra-processed foods is rising globally and has been implicated in inflammation and metabolic dysfunction, yet the impact of specific food additives on the human gut microbiota remains poorly understood. Using dietary data from the Food & You study (approximately 1000 participants in Switzerland), we identified 257 unique additives from 4,119 unique packaged products to quantify each participant's daily additive exposure. Higher exposure to a combination of high intensity sweeteners and sugar polyols, commonly found in low calorie products, was independently associated with reduced gut microbial Shannon diversity (beta = -0.39, p < 0.001), after adjustment for demographics, diet quality, BMI and bowel movement frequency. At a broader level, total additive exposure and fast food consumption were each negatively associated with gut microbial diversity; however, additive exposure remained independently associated and also specifically attenuated the diversity benefits of vegetable rich diets. Furthermore, microbial log ratio signatures linked to additive exposure showed strong negative correlations with Shannon diversity, including emulsifiers and thickeners (r = -0.66) and preservatives and antioxidants (r = -0.56). Integrating additive exposure with healthy dietary components such as HEI, fruits, or vegetables strengthened associations with gut microbial diversity; for example, vegetable linked correlations with Shannon diversity increased from r = 0.52 to r = 0.65 when contrasted against preservative-antioxidant exposure. Concordantly, microbial signatures associated with the sweeteners and sugar polyols additive combination showed depletion of fiber associated commensal taxa, and enrichment of pathways involved in polyol and aromatic compound metabolism. Notably, these associations emerged despite packaged foods representing only approximately 15% of logged dietary intake, underscoring the sensitivity of gut microbial diversity to limited exposure, and demonstrating that without integrating additive and processed-food metrics, one of the largest effect-size phenomena in human gut microbiota diversity would remain undetected.

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

Stochastic Thermodynamics and SDE-based Generative Models

Authors:

arXiv:2606.18290v1 Announce Type: cross Abstract: SDE-based generative models, including diffusion models and the Schrödinger bridge, have found broad applications in signal processing tasks such as speech enhancement, image restoration, and time-series generation. This note presents a modeling framework for such models within the context of stochastic thermodynamics. The main results of this note are trajectory-level definitions of work, heat, and entropy production, along with a generalized Jarzynski identity and a second-law-like inequality. The proposed framework extends the original Jarzynski setup to accommodate time-dependent bath temperature and nonconservative driving forces. This thermodynamic perspective may deepen our understanding of diffusion models and the Schrödinger bridge from a nonequilibrium statistical mechanics viewpoint.

03.
arXiv (CS.CL) 2026-06-24

Self-Recognition Finetuning can Prevent and Reverse Emergent Misalignment

Emergent misalignment (EM) has been linked to the activation of misaligned persona vectors and evil character traits, suggesting that EM operates through disruption of the model's aligned character rather than direct learning of harmful content. Motivated by this connection, we study self-generated text recognition (SGTR) finetuning as a character-targeted intervention that is distinct from existing in-training defenses. We conduct two-stage finetuning experiments across three models (GPT-4.1, Qwen2.5-32B-Instruct, Seed-OSS-36B-Instruct) and multiple EM datasets to compare SGTR finetuning against benign finetuning baselines (correct domain-specific data, general knowledge, and word counting) to find it an effective defense in both reversal and prevention settings. We find that all interventions produce comparable EM reversal, but only when restoring capabilities that EM had degraded. For prevention, only SGTR finetuning consistently reduces misalignment without exacerbating any individual metric, suggesting that character fortification specifically drives prevention. We provide further evidence for EM's relation to the LLM's default character by showing that EM finetuning induces diversity into the LLM's identity self-reports, artificially corrupting self-recognition exacerbates misalignment caused by EM finetuning, and that removing the model's identity-bearing system prompt substantially reduces the effect of EM finetuning. Together, these findings reframe EM not as the adoption of a coherent misaligned persona but as the destabilization of aligned character.

04.
arXiv (quant-ph) 2026-06-24

Compressed Quantum Operators and Roots of Hermite Polynomials

arXiv:2606.24792v1 Announce Type: new Abstract: The fundamental position and momentum operators of quantum mechanics are unbounded, but finite rank compressions of the operators can be considered to obtain partial information on the operators and their properties. Motivated by problems in photonic quantum computing, we bring together results from quantum theory and the theory of orthogonal polynomials to show that a natural finite rank compression of the position and momentum operator representation on Fock space Hilbert space has eigenvalues given by roots of the classical Hermite polynomials. We discuss the corresponding compressed displacement operators and potential applications in approximate quantum error correction.

05.
arXiv (CS.AI) 2026-06-15

Tackling GNARLy Problems: Graph Neural Algorithmic Reasoning Reimagined through Reinforcement Learning

arXiv:2509.18930v3 Announce Type: replace-cross Abstract: Neural algorithmic reasoning (NAR) is a paradigm that trains neural networks to execute classic algorithms by supervised learning. Despite its successes, important limitations remain: inability to construct valid solutions without post-processing and to reason about multiple correct ones, poor performance on combinatorial NP-hard problems, and inapplicability to problems for which strong algorithms are not yet known. To address these limitations, we reframe the problem of learning algorithm trajectories as a Markov decision process, which imposes structure on the solution construction procedure and unlocks the powerful tools of imitation and reinforcement learning (RL). We propose the GNARL framework, encompassing the methodology to translate problem formulations from NAR to RL and a learning architecture suitable for a wide range of graph-based problems. We achieve very high graph accuracy results on several CLRS-30 problems, performance matching or exceeding much narrower NAR approaches for NP-hard problems and, remarkably, applicability even when lacking an expert algorithm.

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

SkyJEPA: Learning Long-Horizon World Models for Zero-Shot Sim-to-Real Control of Quadrotors

arXiv:2606.23444v2 Announce Type: replace-cross Abstract: Accurate dynamics models are critical for informed decision-making in robotic systems, particularly for agile aerial vehicles operating under uncertainty. Neural network dynamics models are attractive for capturing complex nonlinear effects, but existing predictive approaches struggle with long-horizon forecasting because their autoregressive rollout mechanism amplifies errors over time. Joint Embedding Predictive Architectures (JEPAs) offer a compelling alternative by modeling dynamics in latent space, yet prior JEPA-style methods for robot navigation have been studied primarily for kinematic-level planning, with limited investigation in high-frequency control. In this work, we introduce the JEPA-style model for real-time quadrotor control. The proposed approach combines a latent dynamics model with a novel physics-inspired prober that maps frozen latents to interpretable state, enabling physically grounded long-horizon prediction. Additionally, we combine the learned model with a sampling-based optimal control solution to take advantage of its predictive capabilities for real-time control on embedded hardware. Finally, to reduce the dependence on expensive and unsafe real-world data collection, we develop a structured pipeline for automated dataset generation. Extensive open-loop and outdoor closed-loop experiments demonstrate accurate prediction, robust zero-shot sim-to-real transfer, and strong generalization across diverse operating conditions.

07.
arXiv (CS.CV) 2026-06-19

OTCHA: Optimal Transport-driven Confidence-aware Latent Hub Alignment for Multi-View Medical Image Classification

Multi-view imaging, such as mammography and chest radiography, is a standard component of clinical practice. However, medical images are often unregistered and contain view-specific artifacts or irrelevant background cues that can obscure diagnostically relevant findings. Many existing methods directly fuse per-view representations, allowing such irrelevant content to contaminate the fused embedding and reducing robustness under varying view configurations. We propose OTCHA, a confidence-aware latent hub token alignment module based on optimal transport (OT) that refines patch tokens before fusion for multi-view classification. OTCHA introduces a set of learnable latent hub tokens shared across views. For each view, we compute an OT plan between patch tokens and hub tokens that jointly considers feature similarity and geometry, and augment the OT formulation with token-conditional dustbins to enable partial matching and discard irrelevant tokens. The resulting transport plan provides token-wise matching confidence, which gates hub-mediated message passing and weights a novel optimal-transport-based representation alignment loss to stabilize refinement. Experiments on three multi-view medical image datasets demonstrate consistent improvements over competing baselines across diverse anatomies and view configurations. Our code is available at https://github.com/labhai/OTCHA.

08.
bioRxiv (Bioinfo) 2026-06-23

VCBench: A Multi-Dimensional Benchmark for Single-Cell Foundation Models

Single-cell foundation models are increasingly positioned as virtual cells, yet their capabilities are assessed by fragmented, largely single-task benchmarks that obscure where these models improve on simple baselines. VCBench addresses this by synthesizing four independent virtual-cell frameworks into seven capability dimensions: perturbation response prediction, cross-species universality, gene regulatory network (GRN) inference, modality integration, temporal dynamics, multi-scale integration, and in silico experimentation. Each dimension is assessed for operational testability under current architectures and datasets: five admit direct or proxy evaluation, while multi-scale integration and in silico experimentation are structurally untestable as end-to-end tasks. We evaluate five foundation models (Geneformer, scGPT, UCE, TranscriptFormer, Arc State) against pre-registered linear and nearest-neighbor baselines across the five testable dimensions, and report three findings. First, the baselines match or exceed every foundation model on four of the five scored dimensions, replicating the reported competitiveness of linear baselines on perturbation prediction and extending it to cross-species transfer, GRN inference, and temporal ordering. Second, TranscriptFormer alone exceeds the strongest baseline on cross-modal RNA-to-protein prediction (53% Pearson improvement, with a documented contamination caveat) and is the only model to reach Level 2 in the pre-registered Virtual Cell (VC) Level rubric; the architectural choice behind this advantage simultaneously causes a spectral collapse that destroys its temporal-ordering performance, a tradeoff invisible to single-task benchmarks. Third, no foundation model publishes a complete cell-level training manifest, leaving data contamination undetectable to users. Alongside the benchmark, VCBench releases a Contamination Reporting Schema and contributes two further methodological tools: a common-label-set protocol that controls for class-count confounds in cross-species transfer, and a spread-error correlation probe for epistemic calibration.

09.
arXiv (CS.AI) 2026-06-15

Discovery under Hypothesis Redundancy: A Geometric Theory of Discovery Bottlenecks

arXiv:2606.14386v1 Announce Type: cross Abstract: Scientific discovery saturates when new hypotheses cease to provide independent information, even if the nominal hypothesis space remains large. We study hybrid discovery systems that combine structured local search with LLM-generated non-local proposals and pose the Search Compression Hypothesis: non-local exploration helps only when three geometric conditions co-occur: spectral compression, orthogonal escape from the explored span, and residual signal alignment with the target. We formalize these conditions, derive necessary conditions for hybrid advantage, and test the mechanism in controlled synthetic environments, large-scale A-share factor discovery, and symbolic-regression benchmarks; a public tabular operational sanity check tests the associated budget-allocation implication. Signal-planting and directed-versus-random experiments show that novelty alone is insufficient: random orthogonal jumps expand coverage but do not improve yield without predictive alignment. Across compression sweeps, real factor archives, and LLM-SRBench tasks, hybrid gains concentrate in weakly represented but target-bearing directions and vanish as the hypothesis space approaches full rank. The framework turns LLM-guided discovery from generic novelty search into a diagnostic procedure for deciding when directed non-local exploration is warranted.

10.
arXiv (CS.AI) 2026-06-19

Confidence Calibration for Multimodal LLMs: An Empirical Study through Medical VQA

arXiv:2606.19950v1 Announce Type: cross Abstract: Multimodal Large Language Models (MLLMs) show great potential in medical tasks, but their elicited confidence often misaligns with actual accuracy, potentially leading to misdiagnosis or overlooking correct advice. This study presents the first comprehensive analysis of the relationship between accuracy and confidence in medical MLLMs. It proposes a novel method that combines Multi-Strategy Fusion-Based Interrogation (MS-FBI) with auxiliary expert LLM assessment, aiming to improve confidence calibration in Medical Visual Question Answering (VQA). Experiments demonstrate that our method reduces the Expected Calibration Error (ECE) by an average of 40\% across three Medical VQA datasets, significantly enhancing MLLMs' reliability. The findings highlight the importance of domain-specific calibration for MLLMs in healthcare, offering a more trustworthy solution for AI-assisted diagnosis.

11.
arXiv (CS.AI) 2026-06-15

Hyperdimensional computing for structured querying on tabular data embeddings

arXiv:2606.13871v1 Announce Type: new Abstract: Tabular data embeddings have become a cornerstone of data profiling and data integration pipelines, enabling tasks such as entity annotation and resolution; schema matching; column type detection; and table search, among others. Existing approaches embed rows, columns, or entire tables into a vector space and rely on nearest-neighbor search to retrieve candidate matches. A fundamental limitation of current embedding methods is the lack of interpretable similarity scores: the concrete similarity value between a query and its nearest neighbour carries no intrinsic meaning, making it impossible to determine whether that neighbour is a true match or simply the least-dissimilar item in a corpus that contains no valid answer. This inability to set principled thresholds for retrieval undermines practical deployment, particularly for zero-match detection. We investigate the use of HyperDimensional Computing (HDC), specifically the Holographic Reduced Representations (HRR) model, as a framework for tabular row embeddings when the retrieval task corresponds to answering structured select-project queries in vector space. Exploiting the algebraic properties of HDC operations, we derive closed-form expected similarity values for both equality and non-equality retrieval predicates, which converge to interpretable values as dimensionality increases, and use these to identify suitable retrieval thresholds. We evaluate HDC against EmbDI, a graph-based baseline, on two real-world datasets across varying table sizes and predicate lengths. Our results show that HDC matches or outperforms EmbDI for row retrieval across all configurations, handles non-equality predicates more robustly, and achieves perfect attribute projection accuracy at sufficient dimensionality – while uniquely enabling reliable identification of zero-match predicates through its principled thresholds.

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

Multimodal Concept Bottleneck Models

arXiv:2606.19882v1 Announce Type: cross Abstract: Concept Bottleneck Models (CBMs) enhance the interpretability of deep learning networks by aligning the features extracted from images with natural concepts. However, existing CBMs are constrained in their ability to generalize beyond a fixed set of predefined classes and the risk of non-concept information leakage, where predictive signals outside the intended concepts are inadvertently exploited. In this paper, we propose Multimodal Concept Bottleneck Model (MM-CBM) to address these issues and extend CBMs into CLIP. MM-CBM utilizes dual Concept Bottleneck Layers (CBLs) to align both the image and text embeddings into interpretable features. This allows us to perform new vision tasks like zero-shot classification or image retrieval in an interpretable way. Compared to existing methods, MM-CBM achieves up to 51.26% accuracy improvement on average across four standard benchmarks. Our method maintains high accuracy, staying within ~5% of black-box performance while offering greater interpretability.

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

UniECG: Understanding and Generating ECG in One Unified Model

Electrocardiogram (ECG) interpretation is a fundamental skill in medical education, yet students often need more than static examples to connect waveform evidence with diagnostic reasoning. This paper presents UniECG as a step toward interactive ECG education. UniECG supports two complementary learning interactions: given an ECG signal or image, it generates an evidence-based explanation; given a textual learning objective, it generates a corresponding ECG signal example for case-based learning. The model follows a two-stage design. First, it learns grounded ECG explanation from ECG signal–image–text data. Second, it introduces special ECG generation tokens and aligns their hidden representations with a pretrained text-conditioned ECG diffusion model, enabling controllable signal-level ECG generation. We evaluate UniECG through grounded ECG explanation and generation-oriented qualitative analysis, examining its potential to support explanation and case-based learning. UniECG is intended as an educational aid and a research step toward interactive AI-assisted ECG learning, rather than a clinically validated diagnostic system.

14.
arXiv (quant-ph) 2026-06-19

Thermodynamic Value of XOR-Game-Induced Side Information in a Szilard Engine

arXiv:2605.12044v3 Announce Type: replace Abstract: We introduce a Szilard-type thermodynamic valuation of side-information channels induced by Bell-type correlations. In each round, a two-level working system is thermalized with a degenerate Hamiltonian, so that its physical microstate is a uniform classical bit. A trusted referee embeds this bit into a finite two-player XOR game, and a correlation resource produces a compressed controller bit. The controller uses only this compressed bit as side information for feedback. The construction is formulated first for arbitrary finite XOR games. The referee encoding makes the game-winning event equivalent to correct prediction of the physical microstate. Consequently, the induced side-information channel is binary symmetric, with success probability equal to the XOR-game winning probability of the supplied behaviour. The reversible Szilard feedback value is therefore fixed by the mutual information between the microstate and the controller record. Optimizing over local, quantum, and nonsignalling behaviour sets turns the corresponding game values into local, quantum, and nonsignalling thermodynamic ceilings. The construction is an effective-channel valuation, not a claim that Bell nonlocality is thermodynamic fuel. The controller receives only the compressed prediction bit, not the auxiliary variables that define the game. The thermodynamic costs of the referee, the correlation resource, and the preprocessing are not included. When controller-memory reset is included in a full cycle, the net work is non-positive, consistently with the second law.

15.
Nature (Science) 2026-06-17

Molecular basis of polyadenylated RNA fate determination in the nucleus

Authors:

Eukaryotic genomes generate a plethora of polyadenylated (pA+) RNAs1,2, which are packaged into ribonucleoprotein particles (RNPs). To ensure faithful gene expression, functional pA+ RNPs, including protein-coding RNPs, are exported to the cytoplasm, whereas transcripts within non-functional pA+ RNPs are degraded in the nucleus1–4. How cells distinguish these opposing fates remains unknown. The DExD-box ATPase UAP56 (also known as DDX39B) is a central component of functional pA+ RNPs, and promotes their docking to the nuclear pore complex-anchored TREX-25,6, which triggers transcript release from UAP56 to facilitate export7. Here we reveal that the poly(A) tail exosome targeting (PAXT) connection8 binds a TREX-2-like module, which releases pA+ RNAs from UAP56 for decay by the nuclear exosome. The core of this module consists of a LENG8–PCID2–SEM1 trimer, which we show is structurally and biochemically equivalent to the central GANP–PCID2–SEM1 trimer of TREX-2. Mutagenesis and transcriptomic data demonstrate that the nuclear fate of pA+ RNPs is governed by the contending actions of nucleoplasmic PAXT and nuclear pore complex-associated TREX-2, which interpret RNA-bound UAP56 as a signal for RNA decay or export, respectively. As RNA targets of PAXT are generally short and intron-poor, we propose an overall model for pA+ RNP fate determination whereby the distinct sub-nuclear localizations of PAXT and TREX-2 govern the degradation of short non-functional pA+ RNAs while allowing export of their longer and functional counterparts. Biochemical, structural&nbsp;and cell biological analyses reveal that UAP56 (DDX39B) assembles with a TREX-2–like module&nbsp;that redirects non-functional polyadenylated RNAs from export to degradation.

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

OccAny: Generalized Unconstrained Urban 3D Occupancy

Relying on in-domain annotations and precise sensor-rig priors, existing 3D occupancy prediction methods are limited in both scalability and out-of-domain generalization. While recent visual geometry foundation models exhibit strong generalization capabilities, they were mainly designed for general purposes and lack one or more key ingredients required for urban occupancy prediction, namely metric prediction, geometry completion in cluttered scenes and adaptation to urban scenarios. We address this gap and present OccAny, the first unconstrained urban 3D occupancy model capable of operating on out-of-domain uncalibrated scenes to predict and complete metric occupancy coupled with segmentation features. OccAny is versatile and can predict occupancy from sequential, monocular, or surround-view images. Our contributions are three-fold: (i) we propose the first generalized 3D occupancy framework with (ii) Segmentation Forcing that improves occupancy quality while enabling mask-level prediction, and (iii) a Novel View Rendering pipeline that infers novel-view geometry to enable test-time view augmentation for geometry completion. Extensive experiments demonstrate that OccAny outperforms all visual geometry baselines on 3D occupancy prediction task, while remaining competitive with in-domain self-supervised methods across three input settings on two established urban occupancy prediction datasets. Our code is available at https://github.com/valeoai/OccAny .

17.
Nature (Science) 2026-06-10

A prognostic human brain network for diffuse midline glioma

Authors:

Diffuse midline gliomas (DMGs) are near-universally lethal tumours of the&nbsp;childhood central nervous system1,2. In animal models, DMGs form brain-wide integrated networks through neuron-to-glioma synapses3–6 and glioma-to-glioma gap junctional coupling3. This extensive connectivity robustly promotes the growth and invasion of DMG3–9 and other glial malignancies10–12 through paracrine mechanisms and direct neuron-to-glioma synapses. However, the organization and clinical implications of these connections in the living human brain remain to be elucidated. Here, we develop tumour network mapping to compute the brain-wide connectivity profile of DMG, defining a conserved brain network across pontine and thalamic DMG associated with patient short-term survival (DMG network). Tumour functional connectivity with the DMG network was independently predictive of patient overall survival across two external validation cohorts. Tumour growth mapped to DMG&nbsp;network-specific trajectories and peak in-network neurometabolic changes across development spatiotemporally aligned with the peak age incidence of DMG. Analyses of single-nucleus RNA&nbsp;sequencing data&nbsp;confirmed diverse synaptic gene enrichment in high-connectivity DMG. Strikingly, incidental surgical resection of high-connectivity thalamic DMG tissue conferred a significant survival advantage. Collectively, these data define a conserved and prognostically important brain network in children with DMG, consistent with the hypothesis that DMGs exploit otherwise healthy brain circuits to promote tumour growth. Tumour network mapping of diffuse midline glioma&nbsp;(DMG) defines a conserved and prognostically important brain network in children with DMG, consistent with the hypothesis that DMGs exploit otherwise healthy brain circuits to promote tumour growth.

18.
arXiv (CS.LG) 2026-06-24

Sesame: Structure-Aware Molecular Generation via Spatial Density-Map Conditioning

arXiv:2606.23856v1 Announce Type: new Abstract: Generative molecular models for drug design are a promising direction with much active research. In the next phase of computational drug design, such models will need to understand small molecule structure and protein-ligand interactions, and they will need to possess the machinery to generate molecules de novo. Incorporating each feature poses a critical challenge. Equally important, yet often treated as secondary, is the ability to grow a molecule from a partial starting point – a scaffold or fragment supplied by a chemist – which is the central operation of lead optimization. We present Sesame (Spatial Evoformer for a Structure-Aware Molecular Engine), a diffusion-based molecular generation model that leverages a novel spatial pairformer module to condition on partial molecular structure and the surrounding protein pocket, both expressed as continuous spatial density maps. This single conditioning mechanism supports both de novo generation and fragment-conditioned lead optimization, letting a medicinal chemist prune a hit to a scaffold and have Sesame grow it in productive ways. In addition to this module, we also introduce a diffusion framework for joint denoising of atom types, bond types, and positions, along with a trajectory finetuning scheme that trains on the model's own sampling rollouts to improve generation quality. Sesame is trained on a large corpus of ligand-only and protein-ligand datasets.

19.
arXiv (CS.AI) 2026-06-17

Temporal Preference Optimization for Unsupervised Retrieval

arXiv:2606.17664v1 Announce Type: cross Abstract: Unsupervised dense retrievers offer scalability by learning semantic similarity from unlabeled documents via contrastive learning, but they struggle to capture the temporal relevance, retrieving semantically related but temporally misaligned documents-an important aspect when a document collection spans multiple time periods (e.g., retrieving documents from 2018-2025 for "Who is the president in 2019?" introduces temporal ambiguity). Existing methods rely on supervised training with explicit timestamps, which are not always feasible. We propose TPOUR (Temporal Preference Optimization for Unsupervised Retriever), which uses our novel training method Temporal Retrieval Preference Optimization (TRPO). TRPO reinterprets preference learning in the temporal dimension, guiding the retriever to favor temporally aligned documents. TPOUR further generalizes to unseen time periods via interpolation in a learned time embedding, enabling continuous temporal alignment. Experiments on temporal information retrieval (T-IR), TPOUR outperforms both unsupervised and supervised baselines. Compared to Qwen-Embedding-8B, despite being about 72.7x smaller, TPOUR Contriever improves average nDCG@5 by +4.04 (+12.15%) on explicit and +4.98 (+15.21%) on implicit queries. We provide our code at https://github.com/agwaBom/TPOUR.

20.
arXiv (CS.CL) 2026-06-15

An Empirical Study of Automating Agent Evaluation

Agent evaluation requires assessing complex multi-step behaviors involving tool use and intermediate reasoning, making it costly and expertise-intensive. A natural question arises: can frontier coding assistants reliably automate this evaluation process? Our study shows that simply prompting coding assistants is insufficient for this task. Without domain-specific evaluation knowledge, frontier coding assistants achieve only a 30% execution success rate and produce over-engineered evaluations averaging 12+ metrics per agent, indicating that strong coding ability does not automatically translate to reliable agent evaluation. We introduce EvalAgent, an AI assistant that automates the end-to-end agent evaluation pipeline. EvalAgent encodes evaluation domain expertise as evaluation skills (procedural instructions, reusable code and templates, and dynamically retrieved API documentation) that compose into a trace-based pipeline producing complete evaluation artifacts including metrics, executable code, and reports. To systematically assess generated evaluations, we introduce a meta-evaluation framework alongside AgentEvalBench, a benchmark comprising 20 agents, each paired with evaluation requirements and test scenarios. We further propose the Eval@1 metric to measure whether generated evaluation code both executes and yields meaningful results on the first run. Our experiments show that EvalAgent produces focused evaluations, improving Eval@1 from 17.5% to 65%, and achieving 79.5% human expert preference over baseline approaches. Further ablation studies show that evaluation skills are critical for handling complex evaluation: removing them causes Eval@1 to drop significantly from 65% to 30%.

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

Managing Task Execution for Unknown Workloads in Batteryless IoT: A Hardware-Agnostic Evaluation

arXiv:2606.24340v1 Announce Type: new Abstract: In recent years, the Internet of Things (IoT) paradigm has been shifting toward batteryless, energy-harvesting architectures. Sustaining reliable operation in these systems requires intelligent management of highly volatile stored energy. As edge applications grow in complexity, traditional energy-aware schedulers struggle with unpredictable workloads due to their reliance on static execution thresholds or pre-measured, hardware-specific task profiles. To overcome this, we propose two novel, hardware-agnostic dynamic scheduling strategies treating applications as a "black box," requiring no prior energy information: a model-free Reinforcement Learning (RL) agent and an on-the-fly Approximated Prediction (AP) method. We evaluate these methods against an adaptive task rate approach (AsTAR) and optimized static thresholds using a custom-built, physically accurate simulation framework driven by real-world solar data and dynamic LoRa transmission profiles. Rather than claiming universal superiority, our analysis exposes the distinct operational trade-offs of each method: the AP approach delivers lightweight, near-oracle task throughput; the RL agent provides tunable survival-execution balancing; and AsTAR excels at execution pacing across long energy gaps. Finally, we demonstrate that while these advanced strategies provide critical resilience for severely constrained systems with small capacitors, devices with larger energy buffers can efficiently rely on simpler, less computationally expensive static policies.

22.
arXiv (CS.AI) 2026-06-16

Understanding Diversity Collapse in RLVR via the Lens of Overtraining

arXiv:2606.15455v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards (RLVR) has become a key approach for enhancing the reasoning abilities of large language models. However, RLVR often suffers from diversity collapse: Pass@$1$ improves while high-$k$ Pass@$k$ degrades, which is viewed as a narrowing of the model's reasoning boundary. We formalize this diversity collapse through the lens of overtraining: once a problem's contribution to the reference metric has effectively saturated, further updates no longer expand what the model can solve but still concentrate probability mass on the trajectories favored by on-policy sampling. Under a standard setup with few rollouts per problem, even a single observed success places a problem in a nearly saturated regime for high-$k$ Pass@$k$, so most updates in standard RLVR are overtraining from the boundary perspective. This perspective also suggests a reading of whether RLVR can expand the model's reasoning abilities beyond the base model: since RLVR is structurally biased against high-$k$ Pass@$k$, its aggregate decline does not by itself mean that no new reasoning gains occurred. Interventionally, restricting updates to problems with zero observed success lifts Pass@$256$ above the base model on difficult benchmarks; observationally, a non-trivial fraction of initially unsolvable problems become solvable during standard RLVR training. Building on these findings, we propose Bayesian Boundary Gating (BBG), which redirects optimization away from overtraining by estimating each problem's marginal contribution to the reasoning boundary. Across multiple reasoning benchmarks, BBG improves average Pass@$k$ across a wide range of $k$.

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

QK-Normed MLA: QK normalization without full key caching

Query-key (QK) normalization stabilizes attention by controlling the scale of queries and keys before the dot product, but is not immediately compatible with Multi-head Latent Attention (MLA). MLA achieves efficient decoding by caching low-dimensional latent states instead of full keys, whereas post-projection QK RMSNorm appears to require the fully projected key for every cached token. We show this apparent incompatibility is an implementation artifact, not an architectural constraint. RMSNorm decomposes into a static affine weight and a dynamic scalar RMS statistic. The static key-side weight can be absorbed into the MLA query-side projection; the dynamic key statistic reduces to one inverse-RMS scalar per token and KV group. The resulting formulation is exactly equivalent to explicit post-projection QK RMSNorm in exact arithmetic and preserves MLA's latent decode path. In our 400M runs trained for up to 100B tokens, QK-Normed MLA achieves lower training loss and better downstream accuracy than QK clipping, while H800 decode benchmarks show less than 2% latency overhead up to 256k context. These results make QK normalization a practical stabilization option for MLA models without requiring full-key caching.

24.
arXiv (CS.CL) 2026-06-24

A Synthetic Reliability-Aware PINN Benchmark for Offshore Wind Turbine Support-Structure Monitoring with Bayesian Inverse Identification

Reliable structural health monitoring (SHM) of offshore wind turbine (OWT) support structures requires fast state estimation from sparse measurements. Repeated high fidelity finite element or aeroelastic analyses are difficult to use directly in online monitoring loops, while purely data-driven surrogates can require large training sets. This paper presents Digi Turbine, a synthetic reliability-aware Physics Informed Neural Network (PINN) benchmark for OWT monopile support structure monitoring. The workflow embeds a simplified Euler Bernoulli beam equation with Winkler soil foundation in the training objective, couples it with Bayesian-prior-informed inverse identification, and adds First Order Reliability Method (FORM) screening. All validation uses synthetic configurations with analytical or finite-difference ground truth motivated by the NREL 5MW reference turbine context.

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

FinSTaR: Towards Financial Reasoning with Time Series Reasoning Models

arXiv:2605.03460v3 Announce Type: replace Abstract: Time series (TS) reasoning models (TSRMs) have shown promising capabilities in general domains, yet they consistently fail in the financial domain, which exhibits unique characteristics. We propose a general 2 x 2 capability taxonomy for TSRMs by crossing 1) single-entity vs. multi-entity analysis with 2) assessment of the current state vs. prediction of future behavior. We instantiate this taxonomy in the financial domain-where the distinction between deterministic assessment and stochastic prediction is particularly critical-as ten financial reasoning tasks, forming the FinTSR-Bench benchmark based on S&P stocks. To this end, we propose FinSTaR (Financial Time Series Thinking and Reasoning), trained on FinTSR-Bench with distinct chain-of-thought (CoT) strategies tailored to each category. For assessment, which is deterministic (i.e., computable from observable data), we employ Compute-in-CoT, a programmatic CoT that enables models to derive answers directly from raw prices. For prediction, which is inherently stochastic (i.e., subject to unobservable factors), we adopt Scenario-Aware CoT, which generates diverse scenarios before making a judgment, mirroring how financial analysts reason under uncertainty. The proposed method achieves 78.9% average accuracy on FinTSR-Bench, substantially outperforming LLM and TSRM baselines. Furthermore, we show that the four capability categories are complementary and mutually reinforcing through joint training, and that Scenario-Aware CoT consistently improves prediction accuracy over standard CoT. Code is available at https://github.com/seunghan96/FinSTaR.