Academic Intelligence · Curated Daily

探索全球前沿学术脉络

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

01.
Nature Medicine 2026-06-11

Clinical Profile and Genomic Characterization of the 2026 Bundibugyo Virus Index Case in Uganda

Bundibugyo virus disease (BVD) remains a high-consequence threat in Eastern and Central Africa, where cross-border mobility, nonspecific early symptoms, and delayed recognition can obscure transmission. In this case report, we describe Uganda’s 2026 BVD index case: a male patient who traveled from the Democratic Republic of the Congo to Uganda and was admitted to a private hospital in Kampala on 11 May 2026 after more than two weeks of vomiting and diarrhea, with epigastric pain, weakness, and hiccups. He deteriorated rapidly, developing acute kidney injury, pulmonary edema, hepatic dysfunction, hypoxemia, delirium, atrial flutter, possible disseminated intravascular coagulation, and multiorgan failure, and died on 14 May. A posthumous EDTA whole-blood specimen tested at the Central Emergency Response and Surveillance Laboratory was positive for orthoebolavirus RNA and confirmed as Bundibugyo virus (BDBV) by RT-qPCR. Sequencing achieved 99% genome coverage at ≥100× depth. The 2026 BDBV genome formed a distinct lineage approximately equidistant from the 2007–2008 Butalya and 2012 Isiro variants, differing by 216–227 nucleotides (~1.2% sequence divergence). Here, we demonstrate the value of fatality surveillance, private-sector surveillance, diagnostic optimization through national specimen referral, and rapid molecular-genomic diagnostics for early detection, transmission chain interruption, and public health response coordination.

02.
arXiv (CS.LG) 2026-06-16

On the Benefits of Weight Normalization for Overparameterized Matrix Sensing

arXiv:2510.01175v2 Announce Type: replace Abstract: While normalization techniques are widely used in deep learning, their theoretical understanding remains relatively limited. In this work, we establish the benefits of (generalized) weight normalization (WN) applied to the overparameterized matrix sensing problem. We prove that WN with Riemannian optimization achieves linear convergence, yielding an exponential speedup over standard methods that do not use WN. Our analysis further demonstrates that both iteration and sample complexity improve polynomially as the level of overparameterization increases. To the best of our knowledge, this work provides the first characterization of how WN leverages overparameterization for faster convergence in matrix sensing.

03.
arXiv (CS.LG) 2026-06-16

High-Dimensional Random Projection for Activation Steering in Language Models

arXiv:2606.15092v1 Announce Type: new Abstract: Activation steering has emerged as a key methodology for controlling the behavior of large language models (LLMs). Existing difference-in-means based methods, however, are fundamentally limited: they capture only mean differences between class activations and fail to recover discriminative signals that naturally exist in the nonlinear feature subspace under the superposition hypothesis. Motivated by that, we propose High-Dimensional Random-projection for Activation Steering (HiDRA), a training-free approach that integrates seamlessly with existing activation steering methods. By performing activation addition in the projected high-dimensional space, HiDRA can provably capture a better discriminative structure beyond the reach of linear methods. Experiments across diverse LLM families and benchmarks demonstrate that HiDRA consistently outperforms baseline counterparts, achieving stronger behavioral control without significant computational overhead.

04.
Nature (Science) 2026-06-10

Light-induced quantum friction of carbon nanotubes in water

Friction slows down moving objects at both macroscopic and microscopic scales1. At the electronic level, quantum friction describes direct transfer of momentum between a liquid and the electrons of a solid2. Owing to its microscopic nature, this phenomenon remains experimentally challenging to capture3. Here we show that near-infrared fluorescent single-walled carbon nanotubes (SWCNTs) exhibit light-induced quantum friction in water. It is measured by observing an excitation-power-dependent linear decrease of around 50% in the diffusion constants of functionalized SWCNTs in aqueous solution. This effect disappears when excitons are localized, as in the case of SWCNTs with quantum defects. We further show that the chemical manipulation of exciton concentration by molecules that increase or decrease SWCNT fluorescence also modulates the diffusion constant by up to a factor of 2. Optical pump terahertz (THz) probe spectroscopy shows an instantaneous response (around 30 cm−1) that we assign to direct exciton–water coupling in the range of water Debye modes. It is followed by an increasing (>100 ps) response in the range of intermolecular translational modes of the hydrogen bond network of water (>100 cm−1), resembling heating. Classical molecular dynamics simulations further support a mechanism in which the fluctuating dipole moments of excitons create frictional forces. These findings establish light-induced quantum friction between excitons in SWCNTs and water and show that electronic excitations can be used to control nanoscale motion and fluid properties. Near-infrared fluorescent carbon nanotubes exhibit light-induced quantum friction in water, in which exciton interactions slow nanoscale motion and enable optical control of diffusion and fluid dynamics.

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

Rigel: Reverse-Engineering the Metal 4.1 Tensor Compute Path on the Apple M4 Max GPU

Apple's Metal 4.1 exposes a tensor compute path: the Metal Performance Primitives (MPP) matmul2d operation over cooperative_tensor fragments, whose interface is documented but whose hardware behavior is deliberately hidden. The specification states which data-type rows are supported, never whether they are hardware-accelerated, where the operation physically executes, what its accumulator width is, or how it partitions matrix fragments across threads. We present Rigel, an empirical characterization of this path on a single Apple M4 Max (a pre-neural-accelerator generation). Using a checksum-gated, provenance-tracked microbenchmark harness, Rigel recovers eleven facts the v4.1 specification hides or contradicts. The headline finding: the Metal 4.1 fp8 (E4M3) matmul2d is emulated, not accelerated: it sustains 0.94x the throughput of fp16 despite reading half the operand bytes, so on M4 it is a memory-footprint feature, not a performance feature. We further show, via a three-signal triangulation (throughput ceiling, comparison against simdgroup_matrix, and per-rail power attribution), that matmul2d executes entirely on the GPU shader cores with no dedicated matrix datapath and no evidence of Apple Neural Engine routing; that it accumulates in >=fp32; and we reconstruct the opaque 8x8 cooperative_tensor fragment layout Apple documents nowhere. Acting on the characterization, a hand-fused GEMM + bias + GELU kernel beats the decomposed path by +6.5-12.9% in the cache-resident regime. All findings are reproducible from committed MIT-licensed code and per-cell CSVs.

06.
medRxiv (Medicine) 2026-06-23

Respiratory support with Continuous Positive Airway Pressure in preterm neonates: an analysis of coverage and quality of care in 66 neonatal units in Kenya, Malawi, Nigeria and Tanzania implementing with the NEST360 Alliance

Background: Prematurity is the leading cause of child deaths worldwide, with the highest neonatal mortality in sub Saharan Africa. Respiratory distress syndrome (RDS) is the leading mortality pathway in preterm neonates, but continuous positive airway pressure (CPAP) has high impact. This analysis reports CPAP coverage and quality of care for preterm neonates admitted to 66 neonatal units in Kenya, Malawi, Nigeria and Tanzania. Methods: Analyses used individually linked neonatal inpatient data and cross-sectional health systems data. All admitted neonates were eligible for inclusion (January 2021 through December 2024). Service readiness for CPAP delivery and mean CPAP coverage were described for CPAP eligible newborns (weighing 1500g). Quality of care cascades were constructed to illustrate key indicators. Survival among CPAP eligible neonates was analysed using regression models, stratified by clinical severity scores. Results: 375,255 newborn admissions were analysed in 66 neonatal units. Functional CPAP availability varied with median 16% of days (IQR: 4 to 47%) classified as high demand (>1.5 eligible newborns per CPAP). Of 64,761 CPAP eligible neonates, 22,006 (34%, 95% CI 33 to 34%) received CPAP. All countries showed improvement in CPAP coverage, with Tanzanian hospitals recording 63% increase in mean coverage (p-value=0.001) over time. Quality of care cascades showed treatment was initiated 1 day for 42% (95% CI 41 to 43%) of eligible neonates receiving CPAP. Only 10% of neonates

07.
arXiv (CS.AI) 2026-06-18

QSignAI: Quantum-Randomness-Seeded Identity Signatures at the Intersection of AI for Science and Science for AI

arXiv:2605.27729v2 Announce Type: cross Abstract: The 2024-2025 Nobel and Turing awards recognised AI and quantum science simultaneously. Yet no deployed system has brought these streams together for the public. This paper presents QSignAI, a production-deployed platform demonstrating a bidirectional AI-quantum relationship in a real-time event participation system. We address three questions: can quantum-randomness generation via a two-source extractor be embedded in an AI-driven social platform with acceptable latency; can an AI bot make quantum phenomena perceptually legible to general audiences; and does the combined system work in practice? A conversational bot routes each participant's first message through a quantum pipeline comprising a Toeplitz two-source extractor over independent single-qubit Hadamard measurements on SV1 and DM1 simulators, plus a 2-qubit Bell state, producing a unique quantum-randomness-seeded identity signature per participant. The first two questions are answered through system architecture and qualitative deployment evidence from live events; the third through successful production deployment. The current deployment uses cloud quantum simulators; physical QPU randomness is the near-term extension. Measurable benchmarks are identified as priority future work.

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

Stringalign: Moving beyond summary statistics with a transparent Unicode-aware tool for evaluating automatic transcription models

Comparing text strings is crucial when evaluating and understanding the performance of various text processing tasks such as document recognition and audio transcription. With an increasingly complex landscape of AI-based handwritten text recognition (HTR), optical character recognition (OCR) and automatic speech recognition (ASR) models, there is a need for tools that facilitate evaluation in a flexible and reproducible way. This paper presents Stringalign, a Python library designed to simplify the evaluation process for automatic transcription projects and facilitate transparent evaluation. Stringalign's tools to examine and visualise both the rate of errors and the types of errors a model makes, give insights into possible improvements and help inform model selection for a particular task. Widely used string comparison metrics, such as the character and word error rates (CER and WER), although useful, can be ambiguous due to varying definitions of what constitutes a character and a word. Stringalign addresses this challenge by ensuring all preprocessing (i.e. normalisation and tokenisation) is transparent and easily replicable, and by providing tools to move beyond summary statistics and analyse common model errors. Moreover, Stringalign adheres to FAIR (Findable, Accessible, Interoperable, and Reusable) principles for research software while staying lightweight and easy to adapt into researchers existing workflows. In this paper, we discuss challenges with character and word level string comparisons and show through examples that where existing tools can yield opaque and sometimes confusing results, Stringalign provides an easy-to-use and unambiguous alternative.

09.
arXiv (CS.CL) 2026-06-11

Semantic Grading of Written Answers in Low-Resource Language Bangla Using a Fine-Tuned Lightweight Language Model

Bangla is among the world's most widely spoken languages, yet it remains underserved in educational NLP research. In many remote and rural regions, access to qualified subject teachers is limited, and written answers are consequently graded largely by hand, restricting timely and consistent feedback. Automatic assessment is challenging because semantically correct responses can vary substantially in surface form. We present a bilingual (Bangla-English) evaluation system designed for low-resource educational settings that prioritizes semantic correctness over lexical overlap. Our approach fine-tunes a lightweight language model to grade each response using the question, reference answer, and student answer, producing a numeric score and concise, context-grounded feedback suitable for classroom deployment. We also construct a synthetic bilingual dataset to enable controlled training and evaluation. Across proprietary and open-source LLMs evaluated under a unified protocol, our QLoRA-tuned Qwen3-8B confirms consistent improvement by producing the most leakage-resistant feedback (RoRa = 0.819) in synthetic evaluation and the strongest agreement with human scores (rho = 0.936, MAE = 0.725) in a dedicated human study.

10.
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.

11.
arXiv (CS.LG) 2026-06-16

Elastic ODYN: Differentiable Optimization for Infeasible Control and Learning in Robotics

arXiv:2606.16564v1 Announce Type: cross Abstract: Robotic systems routinely encounter conflicting objectives, modeling errors, and degenerate contact conditions that render quadratic programs (QPs) infeasible. Yet most optimization solvers and differentiable QP layers assume feasibility, leading to numerical failures, unstable gradients, or solver breakdown when constraints cannot be simultaneously satisfied. We present Elastic ODYN, a primal–dual non-interior-point QP solver that handles infeasibility through smooth squared-$\ell_2$ elastic relaxations. The resulting formulation remains well posed under ill-conditioning and degeneracy, supports warm starting, and converges to closest-to-feasible solutions when no feasible point exists. A lightweight refinement stage recovers physically meaningful dual variables from the elastic solution. Building on this framework, we develop Elastic OdynLayer, a differentiable QP layer with stable gradients under infeasibility, and Elastic OdynSQP, an infeasibility-aware SQP method that resolves inconsistent subproblems and intrinsically infeasible optimal control tasks through selective constraint relaxation. We evaluate the framework on benchmark QPs, singular contact mechanics, differentiable parameter identification, and quadrupedal and humanoid trajectory optimization. Across all settings, Elastic ODYN consistently outperforms state-of-the-art elastic QP solvers in robustness, warm-start performance, and convergence reliability, enabling optimization, simulation, control, and learning beyond the feasibility assumptions of existing methods.

12.
arXiv (CS.CL) 2026-06-17

ReproRepo: Scaling Reproducibility Audits with GitHub Repository Issues

Reproducing research results from papers and released code is central to scientific progress. Existing works have introduced benchmarks to evaluate whether LLM agents can assist with reproducibility, but they are difficult to scale due to their reliance on substantial manual effort for data curation and evaluation. We introduce ReproRepo, a scalable framework for reproducibility evaluation that leverages human-raised GitHub issues as naturally occurring supervision on realistic reproduction blockers. We instantiate ReproRepo on 1,149 recent machine learning papers from major conferences and evaluate four frontier model-agent configurations. Our results show that LLM agents, even without executing code, can identify many real-world reproducibility problems from paper-repository pairs: the best agent in our study, namely Codex with GPT-5.5, surfaces at least one semantically related human-reported blocker for ~90% of papers in the study. Further analysis shows that agents are particularly effective for surfacing visible failures and identifying the right semantic region, but may still be insufficient in exact localization. ReproRepo can serve as a reusable, scalable framework for future evaluations of LLM agents on real-world reproducibility auditing. Our code is released at https://github.com/LithiumDA/ReproRepo.

13.
medRxiv (Medicine) 2026-06-22

National trends and operational drivers of vaccine wastage in Uganda, 2020-2025: a descriptive analysis of four tracer antigens

Background Vaccine wastage reduces immunisation efficiency, increases costs, and complicates supply forecasting. Uganda routinely monitors vaccine use, but national evidence comparing observed wastage with World Health Organization (WHO) and Uganda-specific planning thresholds has been limited. We described national and sub-national trends for four tracer antigens to inform supply-chain planning and forecasting. Methods We conducted a retrospective descriptive analysis of routinely reported immunisation data from Ugandas District Health Information Software 2, 2020-2025. We analysed Bacille Calmette-Guerin (BCG), measles-rubella (MR), oral polio vaccine (OPV), and diphtheria-tetanus-pertussis-containing vaccine (DPT). Vaccine wastage was calculated as the proportion of issued doses not administered. Annual wastage rates were summarised using medians, and temporal trends were assessed using the Mann-Kendall test. Observed wastage was compared with WHO thresholds: BCG[≤]50%, MR[≤]25%, OPV[≤]10%, DPT[≤]15%, and Ugandas planning thresholds: BCG[≤]70%, MR[≤]40%, OPV[≤]15%, DPT[≤]10%. Effective Vaccine Management reports were reviewed to summarise reported reasons for wastage. Results During 2020-2025, median national wastage was 40.6% for BCG, 25.9% for MR, 10.0% for OPV, and 9.2% for DPT. OPV wastage declined from 12.8% in 2020 to 8.0% in 2025, with a significant downward trend ({tau}b=-1.00; p=0.008). OPV and DPT wastage remained largely within their respective Uganda in-country thresholds ([≤]15% and [≤]10%) for most of the study period, while BCG generally remained below the WHO threshold ([≤]50%) and MR frequently exceeded the WHO threshold ([≤]25%) but remained within Uganda's planning threshold ([≤]40%) in most years. The proportion of districts exceeding both WHO and Uganda thresholds declined for OPV from 36.3% to 5.5% (p=0.024) and for DPT from 22.6% to 1.4% (p=0.013). Wastage was consistently higher in lower-level (Health Centre II and III) facilities, compared to hospitals. Among 50 service delivery points, reported reasons included low session attendance (66%), multi-dose vial policy non-compliance (28%), and vaccine expiry (12%). Conclusion Uganda achieved reductions in OPV wastage and district-level improvements in DPT wastage, while BCG and MR remained more variable and frequently had higher wastage. Strengthening adherence to the multi-dose vial policy and improving session planning at lower-level facilities could strengthen vaccine utilisation and forecasting.

14.
arXiv (CS.LG) 2026-06-24

Dirac-Frenkel dynamics with inertia for nonlinearly parametrized solutions of evolution problems

arXiv:2606.24769v1 Announce Type: cross Abstract: Even when Dirac-Frenkel dynamics determine a well-defined evolution in function space, the corresponding parameter dynamics can be non-unique or ill-conditioned for redundant nonlinear parametrizations such as neural networks or mixture models. We propose to add inertia to the Dirac-Frenkel dynamics and show that this allows useful parameter velocity information to persist from the past trajectory in directions that are weakly informed, while well-informed parameter velocity directions continue to follow the Dirac-Frenkel dynamics. We prove that the inertial formulation yields well-posed parameter dynamics and provide a posteriori error bounds. After time discretization, the method requires the solution of the same type of regularized linear least-squares problem as standard Dirac-Frenkel dynamics, but with the previous velocity appearing as an anchor. Numerical experiments demonstrate the increased robustness obtained with inertia.

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

Noise-Aware Framework for Correcting Corrupted Labels

arXiv:2606.11695v1 Announce Type: cross Abstract: High-quality labeled data is essential for training reliable ML/DL models. However, real-world datasets often contain a considerable proportion of corrupted labels, which can severely degrade model performance. To address this problem, we propose CANOLA, a novel framework for correcting corrupted labels through noise-aware learning and iterative label refinement. CANOLA explicitly estimates the underlying noise distribution of the dataset and incorporates this information into the training of a noise-aware Deep Neural Network. By incorporating noise characteristics during learning, CANOLA enables the model to down-weight unreliable supervision signals and focus on trustworthy patterns, thereby improving robustness and generalization. Label correction is performed via cautious, iterative soft label refinement, in which model predictions are blended with observed labels to prevent premature or erroneous updates. This progressive refinement allows the dataset to be repaired in a stable and controlled manner. We evaluate CANOLA on six widely used datasets under realistic noisy labeling scenarios. Experimental results show that CANOLA consistently outperforms SOTA label correction methods, achieving relative improvements ranging from 19% to 52% in error reduction. Moreover, models trained on datasets corrected by CANOLA obtain substantial downstream performance gains. Even simple classifiers trained on CANOLA's corrected data can outperform complex model-centric approaches by margins of up to 67%.

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

Rigorous extension of semilocal collinear functionals to noncollinear DFT using $SU(2)$ rotations

arXiv:2605.31203v2 Announce Type: replace-cross Abstract: In the presence of spin-orbit coupling and in geometrically frustrated materials, a noncollinear treatment the magnetization density is essential. However, in density functional theory most exchange–correlation functional approximations were originally developed for locally collinear magnetization. Many practical approaches to noncollinear DFT have emerged over the past decade. However, a first-principles connection between widely used semilocal collinear functionals and their noncollinear generalizations remains lacking. In this work, a locally exact relation between collinear and noncollinear exchange–correlation functionals is derived at the level of gradient expansions within a $u(2)$ matrix representation of the energy functional. Within this framework, collinear semilocal variables naturally acquire distinct dependencies on transverse and longitudinal magnetization gradient components. The widely used Scalmani–Frisch scheme emerges as a first-order approximation. The transformation of collinear functional derivatives to noncollinear space is implemented through numerically robust $SU(2)$ rotations. A consistent description of local magnetic torques is demonstrated for the prototypical spin-frustrated Cr$_3$ cluster. The approach further extends to fully nonlocal functionals and provides a direct route towards numerically stable relativistic response calculations. The influence on magnetic properties in presence of spin-orbit coupling is illustrated through calculations of hyperfine couplings in the high-spin ground states of uranium and the uranium ion.

17.
arXiv (CS.AI) 2026-06-24

MOCHA: Multi-modal Objects-aware Cross-arcHitecture Alignment

arXiv:2509.14001v5 Announce Type: replace-cross Abstract: Personalized object detection aims to adapt a general-purpose detector to recognize user-specific instances from only a few examples. Lightweight models often struggle in this setting due to their weak semantic priors, while large vision-language models (VLMs) offer strong object-level understanding but are too computationally demanding for real-time or on-device applications. We introduce MOCHA (Multi-modal Objects-aware Cross-arcHitecture Alignment), a distillation framework that transfers multimodal region-level knowledge from a frozen VLM teacher into a lightweight vision-only detector. MOCHA extracts fused visual and textual teacher's embeddings and uses them to guide student training through a dual-objective loss that enforces accurate local alignment and global relational consistency across regions. This process enables efficient transfer of semantics without the need for teacher modifications or textual input at inference. MOCHA consistently outperforms prior baselines across four personalized detection benchmarks under strict few-shot regimes, yielding a +10.1 average improvement, with minimal inference cost.

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

Sentence-Level Contextual Entrainment in Large Language Models

Contextual entrainment, which is a newly discovered phenomenon in large language models (LLMs), refers to the tendency of a model to assign higher probabilities to tokens that appear in its context. In this work, we extend this phenomenon from the token level to the sentence level by examining the per-token mean log-probability of a sentence instead of the probabilities of individual tokens. We investigate sentence-level contextual entrainment across 26 LLMs from seven families and two datasets, which cover both subjective and objective tasks. We find that sentence-level contextual entrainment exists. This means that the sentences in the prompt (even if they are counterfactual statements) can significantly increase their probability during model inference time. As the model size increases, contextual entrainment gradually decreases. We also find that contextual entrainment is controlled by 2% to 4% of the attention heads. Turning off these attention heads can effectively mitigate contextual entrainment without hurting the model's performance.

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

Optimizing Lithium Production Decisions under Geological, Demand, and Pricing Uncertainties: A POMDP Framework for Multi-Objective Decision Making

arXiv:2606.18598v1 Announce Type: new Abstract: Decision making in lithium production is challenging, whether from an investor's perspective or a strategic production standpoint. Determining which mines to open and when to open them involves not only geological and price uncertainties, but also complexities around the choice of extraction method, from direct lithium extraction to hard rock mining. Prior work explored models of this problem and different methods to optimize mining decisions; these models did not account for uncertainty in pricing, uncertainty in demand, or different mining technologies to extract lithium. Incorporating different pricing models and extraction technology into these models enables more robust strategies for determining not only when and where to open a mine, but also which method of production to pursue. We frame the problem as a partially observable Markov decision process (POMDP) and solve using belief state planning methods to get optimal decision making. In our study, we show that POMDP solvers outperform human inspired heuristics by dynamically adapting to shifting lithium price regimes (static, linear, exponential, and stochastic) through belief state planning and explicit uncertainty management. By optimally sequencing exploration, production, and technology choice, the framework achieves higher demand fulfillment and more balanced economic environmental outcomes over the projects lifetime in all different pricing and deposit scenarios.

20.
arXiv (CS.AI) 2026-06-17

SEAGym: An Evaluation Environment for Self-Evolving LLM Agents

arXiv:2606.17546v1 Announce Type: new Abstract: Self-evolving LLM-based agents improve mainly by changing their agent harness: the structured execution layer around a base model, including prompts, memory, tools, middleware, runtime state, and the model-tool interaction loop. Existing evaluations often reduce this process to isolated task scores or a single sequential curve, obscuring whether an update produces reusable improvement, overfits recent tasks, increases cost, or harms older behavior. We introduce SEAGym, an evaluation environment for measuring agent harness updates across training, validation, test, replay, and cost records. SEAGym turns Harbor-compatible benchmarks into dynamic self-evolution task sources with train batches, frozen update-validation, held-out ID and OOD transfer views, replay diagnostics, and saved snapshot and metric records. Instantiating SEAGym on Terminal-Bench 2.0 and HLE, we compare ACE, TF-GRPO, and AHE under a shared epoch/batch protocol. The results show that these evaluation views provide complementary signals about the evolution process: frequent updates may fail to improve held-out performance, useful intermediate snapshots may collapse later, and source diversity and model backend can affect harness reliability.

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

Legal Reasoning Is Not Lawyering: Rethinking Legal Benchmarks for Pro Se Access to Justice

arXiv:2606.23716v1 Announce Type: cross Abstract: Legal AI benchmark research frequently invokes the assumption that large language models can improve access to justice, including for people who cannot access lawyers in order to understand and exercise their legal rights. We argue that current benchmarks are not equipped to support this assumption because they evaluate legal reasoning over inputs that have already been preprocessed by legal experts, which measures the upper bound of model performance. Access to justice depends on a lower bound: how models perform when inputs come from pro se litigants, whose prompts may contain noisy narratives, buried facts, omissions, folk-legal assumptions, and surface-level errors. These degradations are comparable to conditions under which LLMs are known to degrade in the general machine learning literature, including long-context sensitivity, underspecification, hallucination, and typographical perturbations. We connect evidence from pro se literature with this body of machine learning research and present a small perturbation experiment on LEXam, a legal benchmark, to illustrate the gap between these two bounds. If model development continues to focus on benchmarks that measure only the upper bound, this gap may remain hidden or even widen. We conclude by calling for legal benchmarks that directly measure robustness under pro se-like inputs so that access-to-justice claims about legal AI can become empirically testable.

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

DeFAb: A Verifiable Benchmark for Defeasible Abduction in Foundation Models

arXiv:2606.18557v1 Announce Type: new Abstract: A rule-based logic solver resolves every instance in our benchmark in under 50 microseconds with 100% accuracy; the best frontier language model reaches 65% at best and drops to 23.5% under rendering-robust evaluation (worst case over four surface renderings). We introduce DeFAb (Defeasible Abduction Benchmark), a dataset and generation pipeline that converts four decades of publicly funded knowledge bases into formally grounded instances for defeasible abduction: constructing hypotheses that explain anomalies by overriding defaults while preserving unrelated expectations. Because every hypothesis must pass polynomial-time checks for valid derivation, conservativity, and minimality, DeFAb makes logical rigor the instrument for measuring creativity and theoretical reasoning, scoring the disciplined construction of theory revisions rather than fluent but theory-destroying prose. The pipeline pairs taxonomic hierarchies (OpenCyc, YAGO, Wikidata) with behavioral property graphs (ConceptNet, UMLS) to produce 372,648+ instances across 33.75M materialized rules from 18 sources, in three levels with polynomial-time verifiable gold standards. Four frontier models do not reliably internalize defeasible reasoning: rendering-robust Level 2 accuracy is 7.8-23.5%; chain-of-thought variance (~36 pp) exceeds any inter-model gap; and a matched contamination control isolates a +19.4 pp Level 3 gap. We further release DeFAb-Hard (a 235-instance Level 3 difficulty variant; best model 53.3% vs 100% symbolic) and CONJURE (a kernel-verified transformative-creativity variant of 560 Lean 4/Mathlib instances whose gold answers are definitions the proof kernel did not previously contain, judge-free verifier; a pilot finds zero novel concepts). The same verifier doubles as an exact reward for preference optimization (DPO, RLVR/GRPO). Released under MIT at https://huggingface.co/datasets/PatrickAllenCooper/DeFAb.

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

LLM4RTL: Tool-Assisted LLM for RTL Generation

arXiv:2606.15500v1 Announce Type: cross Abstract: Large language models (LLMs) have facilitated impressive progress in software engineering, code generation, tooling, and systems. Concurrently, a significant body of research has developed which explores a growing variety of methods and systems for applying LLMs to hardware and chip design (e.g., systems for RTL code generation based on functional description). However, when it comes to open Verilog/RTL code-generation, we need high-quality training samples to build specialized and more effective LLM systems through fine-tuning or low-rank adaptation. Here, we propose a ``judge-renew-check-renew-check'' (JRCRC) pipeline which updates a current public dataset using a hierarchy of state-of-the-art commercial LLM models differing in their costs and capabilities in RTL code generation. This approach achieves a cost-effective mechanism for filtering and refining code-generation samples into a higher-quality training dataset. Our experiments also identify some common weaknesses of LLMs in rule-based reasoning and logic, and consequently, in RTL code-generation. Having identified these weaknesses, we develop an architecture for incorporating pre-processing tools to dynamically assist the LLMs in inferring logical relationships from tabular data formats. With our tools-assisted architecture for RTL code generation, we achieve significant overall performance gains in the VerilogEval benchmark and outperform many state-of-the-art methods. Our LLM4RTL system achieves performance comparable to that of GPT-4O using a significantly much smaller LLM.

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

Benchmarking LLM Agents on Meta-Analysis Articles from Nature Portfolio

Meta-analysis is a demanding form of evidence synthesis that combines literature retrieval, PI/ECO-guided study selection, and statistical aggregation. Its structured, verifiable workflow makes it an ideal substrate for evaluating systematic scientific reasoning, yet existing benchmarks lack ground truth across the full retrieval-screening-synthesis pipeline. We introduce MetaSyn, a dataset of 442 expert-curated meta-analyses from Nature Portfolio journals. Each entry pairs a research question with PI/ECO criteria, a retrieval corpus of 140k PubMed articles, verified positive studies, hard negatives that are topically similar but PI/ECO-ineligible, and complete search strategies and date bounds. Benchmarking twelve pipeline configurations (nine RAG variants and a protocol-driven agent) reveals a critical screening bottleneck: despite a retrieval ceiling of 90.9% recall at K=200, no system recovers more than 52.7% of ground-truth included literature. Current LLMs fail to reliably separate eligible studies from PI/ECO-failing distractors in pools of comparable topical relevance. Stage-attributed metrics capture where systems succeed and fail; a single end-to-end score does not.

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

LLM Features Can Hurt GNNs: Concatenation Interference on Homophilous Graph Benchmarks

Adding LLM-generated node features to graph neural networks (GNNs) is widely reported to improve accuracy on standard benchmarks. We document a contrasting observation: when LLM features are introduced through pure input concatenation (rather than joint training, distillation, or prompt-conditioning), they can systematically degrade accuracy on the same homophilous benchmarks where end-to-end LLM pipelines succeed. With an MLP backbone on the Planetoid public split and bag-of-words original features, concatenating SBERT-encoded GPT-4o-mini TAPE features reduces PubMed test accuracy by -17.0 +/- 0.3 pp and Cora by -4.3 +/- 0.6 pp (CiteSeer -0.6 +/- 0.8 pp, within seed noise). The drop attenuates as we relax each condition (GCN / GCNII / GAT backbones, random splits, smaller encoders) and reverses on medium-homophily WikiCS (+4.4 pp) and ogbn-arxiv (+11.7 pp). To predict when concatenation helps versus hurts, we report a simple measure of LLM-alone discriminability, Delta_sig. Across 9 datasets Delta_sig correlates with the concatenation cost more strongly than homophily at point estimate (r^2 = 0.38 vs. 0.06; N=9, bootstrap CIs overlap). The bootstrap-best change-point is tau = 13.8 pp, and the rule "Delta_sig