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.
medRxiv (Medicine) 2026-06-22

MinderCare: protocol for a mixed-methods evaluation of a digitally enabled dementia care service.

Introduction and aims Dementia is a growing public health challenge affecting millions of people worldwide. It is a progressive condition that increases the risk of infections, falls, hospital admissions, dependence in activities of daily living, safety issues such as wandering, care home transfers, and death. New ways of supporting people living with dementia (PLWD) at home are urgently needed. We describe the MinderCare study which evaluates a digitally enabled care model that integrates low-burden sensor-based remote monitoring within a nurse-led clinical service. Methods and analysis In this mixed-methods study, we will recruit 100 people with confirmed or suspected dementia living at home and deploy the Minder remote monitoring system for at least 12 months. A detailed characterisation of the cohort will be obtained, including cognition, frailty, participant and carer wellbeing, functioning, and quality of life. The feasibility, acceptability, sustainability, and resource requirements of the service will also be assessed. Low-cost sensors provide information about behaviour, environment and physiology from the home. Machine-learning algorithms have been used to develop digital biomarkers of infection, sleep, night-time behaviours, daily activities and routines, and the effects of clinical events and treatment. These will be assessed through clinical reports of sensor-derived data that include anomaly alerts provided to the clinical teams. Algorithms will be assessed for their clinical utility and acceptability. The comparative-effectiveness component will be designed as a target trial emulation using linked electronic health-record data to construct a time-indexed external usual-care control cohort. The primary comparative outcome will be Days Alive and Out of Hospital (DAOH) over 12 months from the activation-index date, with healthcare utilisation, costs, institutionalisation and mortality assessed as secondary outcomes. DAOH and estimated MinderCare effects will also be examined across prespecified strata of baseline inpatient utilisation. Ethics and dissemination Ethical approval has been granted by the North East Newcastle and North Tyneside 2 Research Ethics Committee, and the study has received confirmation of capacity and capability by the Imperial College Healthcare NHS Trust. Study findings will be disseminated to patients, health and social care professionals, and policymakers through peer-reviewed publications and conference presentations. Study registration number: ISRCTN14997677 and NIHR portfolio CPMSID 63023.

02.
arXiv (CS.CV) 2026-06-15

HARBOR: Heading Analysis and Reconstruction from Behavioral Observation and Radar

Maritime situational awareness often relies on Automatic Identification System (AIS) transmissions to track vessel movements. However, in operational or conflict scenarios, these data may be unavailable due to signal loss, deliberate deactivation, or intentional spoofing. In such conditions, synthetic aperture radar (SAR) imagery becomes a critical sensing alternative for wide-area maritime monitoring, despite providing only static scene snapshots. This work introduces HARBOR (Heading Analysis and Reconstruction from Behavioral Observation and Radar), a complete pipeline for transforming a single SAR image into predictive motion information without requiring any auxiliary data source at inference time. The method begins with SAR image preprocessing to enhance and segment vessel candidates, followed by automatic detection, size-based classification, and heading estimation using skeleton geometry and local intensity patterns. AIS data are used exclusively during an offline calibration phase to derive vessel-type-dependent motion parameters, which are then applied to generate probabilistic heatmaps of candidate future vessel positions. A case study using real COSMO-SkyMed SAR imagery demonstrates the pipeline on a maritime scene in southern Brazil, showing its ability to extract motion tendencies and generate probabilistic projections of vessel positions in data-denied environments.

03.
arXiv (CS.LG) 2026-06-11

Bergson: An Open Source Library for Data Attribution

arXiv:2606.11660v1 Announce Type: new Abstract: Data attribution is a promising field in interpretability that aims to explain model behavior through the influence of its training data, with applications including debugging undesirable model behavior and training dataset curation. However, significant engineering effort is required to perform it at scale, and many cutting edge techniques lack open-source tooling and support. Bergson is an open source library that aims to enable faster progress in the field by providing a host of techniques that scale to very large language models and pre-training datasets. The library natively supports on-disk gradient stores and multi-node distributed training, and provides quality of life tools for researchers. Finally, we introduce the first open-source implementations of three leading data attribution methods: MAGIC, SOURCE, and TrackStar. The library is available at https://github.com/EleutherAI/bergson .

04.
arXiv (quant-ph) 2026-06-16

Worst-case depth hierarchy for shallow quantum circuits

arXiv:2606.16425v1 Announce Type: new Abstract: Circuit depth is a central resource in complexity theory. While bounded-depth classical circuits admit well-understood hierarchy theorems, the internal structure of constant-depth quantum computation remains comparatively unexplored. We prove an explicit depth hierarchy theorem for $\mathsf{QNC}^0$. For each $d\ge 12$, we construct a family of two-round interactive problems on which no depth-$(d-1)$ quantum circuit can achieve near-perfect success, regardless of gate set, circuit size, or ancillary qubits. In contrast, we prove that our construction admits realizations by simple bounded fan-in quantum circuits of depth larger than $d$ by a small constant factor. Moreover, all bounded fan-in classical circuits of sublogarithmic depth (in the input size) fail to achieve perfect success on these tasks for every $d$, yielding a hierarchy of problems that show unconditional quantum advantage of $\mathsf{QNC}^0$ over $\mathsf{NC}^0$. A key obstacle is the scarcity of lower bound techniques for quantum circuits. To address this, we develop methods to analyze how depth affects a circuit's ability to realize nonlocal correlations amongst its output qubits in a fine-grained manner. Our approach exploits the correspondence between constraint systems and nonlocal games, translating group-theoretic constructions into rigid operator-valued constraint systems and then into non-local games. In particular, we construct constraint systems whose unique faithful operator-valued solutions require every perfect strategy, and every near-perfect strategy to a fixed precision, to implement multi-controlled phase operations. This reduces to a nonlocal unitary-synthesis problem, yielding depth lower bounds for both shallow quantum and classical circuits. These results show that increasing depth strictly increases computational power within $\mathsf{QNC}^0$, establishing a genuinely quantum hierarchy.

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

Phishing Email Detection Using Large Language Models

arXiv:2512.10104v2 Announce Type: cross Abstract: Email phishing is one of the most prevalent and globally consequential vectors of cyber intrusion. As systems increasingly deploy Large Language Models (LLMs) applications, these systems face evolving phishing email threats that exploit their fundamental architectures. Current LLMs require substantial hardening before deployment in email security systems, particularly against coordinated multi-vector attacks that exploit architectural vulnerabilities. This paper proposes LLMPEA, an LLM-based framework to detect phishing email attacks across multiple attack vectors, including prompt injection, text refinement, and multilingual attacks. We evaluate three frontier LLMs (e.g., GPT-4o, Claude Sonnet 4, and Grok-3) and comprehensive prompting design to assess their feasibility, robustness, and limitations against phishing email attacks. Our empirical analysis reveals that LLMs can detect the phishing email over 90% accuracy while we also highlight that LLM-based phishing email detection systems could be exploited by adversarial attack, prompt injection, and multilingual attacks. Our findings provide critical insights for LLM-based phishing detection in real-world settings where attackers exploit multiple vulnerabilities in combination.

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

Towards Understanding and Measuring COGNITIVE ATROPHY in LLM Behaviour

arXiv:2606.18129v1 Announce Type: cross Abstract: Recent incidents involving LLMs used for mental-health support reveal a critical evaluation gap: surface-level safety scores do not capture how models behave across realistic, emotionally sensitive interactions over time. Existing benchmarks measure knowledge, safety, or static response quality, but miss whether LLM interactions help users keep reflecting, coping, and making decisions themselves. We formalize this missing dimension as COGNITIVE ATROPHY, a process-level behavioural measure in AI-mediated mental-health support distinct from safety and helpfulness. To measure it, we introduce COGNITIVE ATROPHY BENCH, a clinically grounded benchmark built from 1,576 fully human-generated counseling conversations, 15,680 turns, and 42,230 responses from five LLMs. Three clinical and neuropsychology experts developed a 20-attribute schema spanning user context, response behaviour, and global risk flags; six trained clinical reviewers applied it with span-grounded evidence, producing 5,324 reviewer judgments. We further introduce the User-Input Risk Index (UIRI), the Cognitive Atrophy Risk Index (ARI), and trajectory summaries. Across five LLMs, models show a consistent moderate-to-high level of atrophy-aligned behaviour across single and multi-turn settings. While models generally respond to overt safety cues, they adapt less reliably when users seek solutions or decisions. The dominant recurring patterns are directive advice, problem-solving, recommendation responses, topic shifts, and forms of validation that may reinforce dependence rather than reflection. Our work makes COGNITIVE ATROPHY measurable and provides a foundation for auditing model behaviour in sensitive LLM conversations.

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

ExpRL: Exploratory RL for LLM Mid-Training

arXiv:2606.17024v1 Announce Type: new Abstract: Sparse reward reinforcement learning (RL) has become a standard tool for improving LLM reasoning, but its success depends critically on the coverage present in the base model. In practice, models are often primed for RL through mid-training on curated reasoning traces that teach useful primitive skills such as decomposition, verification, or self-correction. Although effective, this strategy requires manually specifying what the model should learn, and it remains unclear whether such primitive coverage is enough for much harder problems, which require combining these skills into broader solution strategies. We study a more automated approach: RL-based mid-training using large corpora of human-written question-answer data. Rather than treating reference solutions as targets to imitate, our method, ExpRL, uses them as reward scaffolds: references are hidden from the policy and used only to construct problem-specific grading rubrics for judging on-policy reasoning traces. The policy samples from the original problem prompt, while an LLM judge compares the sampled reasoning trace against the reference solution and assigns outcome-level or process-level dense rewards. This lets ExpRL reinforce partial progress, useful intermediate reductions, and productive reasoning behaviors that sparse final-answer rewards often fail to upweight. On challenging math reasoning tasks, ExpRL yields stronger RL priming than SFT, sparse-reward GRPO, and self-distillation, and provides a better initialization for subsequent sparse-reward RL. Additional mixed-domain experiments further suggest that ExpRL can extend beyond the original math-only setting.

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

On Regret Bounds of Thompson Sampling for Bayesian Optimization

arXiv:2603.09276v2 Announce Type: replace-cross Abstract: We study a widely used Bayesian optimization method, Gaussian process Thompson sampling (GP-TS), under the assumption that the objective function is a sample path from a GP. Compared with the GP upper confidence bound (GP-UCB) with established high-probability and expected regret bounds, most analyses of GP-TS have been limited to expected regret. Moreover, whether the recent analyses of GP-UCB for the lenient regret and the improved cumulative regret upper bound can be applied to GP-TS remains unclear. To fill these gaps, this paper shows several regret bounds: (i) a regret lower bound for GP-TS, which implies that GP-TS suffers from a polynomial dependence on $1/\delta$ with probability $\delta$, (ii) an upper bound of the second moment of cumulative regret, which directly suggests an improved regret upper bound on $\delta$, (iii) expected lenient regret upper bounds, and (iv) an improved cumulative regret upper bound on the time horizon $T$. Along the way, we provide several useful lemmas, including a relaxation of the necessary condition from recent analysis to obtain improved regret upper bounds on $T$.

09.
arXiv (CS.CV) 2026-06-16

Revealing Artifacts via Noise Amplification: A Novel Perspective for AI-Generated Video Detection

With the rapid advancement of video generation models, distinguishing between AI-generated and authentic videos has emerged as a challenging endeavor. The majority of existing research endeavors concentrate on the development of detectors for identifying samples generated by generative adversarial networks. Nevertheless, the detection of AI-generated videos, particularly those produced by text-to-video models, still remains an uncharted territory. Although state-of-the-art text-to-video models can generate realistic visual content similar to real videos, they fall short of generating the details of the images and the changes in details within the videos. Inspired by this, we address AI-generated video detection from a novel perspective of bit-planes, which can effectively describe the details or noises in images or videos. To this end, we propose a simple yet effective approach called Noise Amplification. This approach first extracts noise signals based on bit-planes, then amplifies these noise signals, and finally feeds them into the discriminator networks for video fake classification. Noise amplification is comprehensively constructed by incorporating three aspects: pixel-level intensity enhancement, region-level spatial amplification, and frame-level temporal aggregation. To evaluate methods of AI-generated video detection in challenging scenarios, we also introduce a benchmark named HardGVD. Extensive experiments on both the large-scale dataset GenVidBench and HardGVD show that our simple approach significantly outperforms state-of-the-art methods.

10.
arXiv (CS.CV) 2026-06-16

SceneCraft: Interactive System for Image Editing via Scene Graph

Recent advances in generative AI have enabled natural language-driven image editing, yet existing systems often fail in complex scenes with multiple interacting objects because they rely heavily on users crafting precise text prompts. To address the absence of structured control, we propose SceneCraft, a novel interactive framework that bridges user intent and model execution by representing images as editable scene graphs. Instead of guessing text prompts through trial and error, users interact directly with a visual graph to perform complex spatial and relational operations. These graph modifications are automatically translated into precise, context-aware editing prompts, effectively eliminating linguistic ambiguity. To ensure robust and diverse results, structured prompts are dispatched to multiple state-of-the-art generative models. Evaluations across diverse editing scenarios show that SceneCraft provides a more intuitive control mechanism, significantly reducing the cognitive burden of manual prompt engineering while generating outputs that users consistently rate as higher in quality and fidelity.

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

Generative Molecular Design with Steerable and Granular Synthesizability Control

arXiv:2505.08774v2 Announce Type: replace-cross Abstract: Designing molecules that are both property-optimal and readily synthesizable is a central challenge in drug discovery. Existing works that do consider synthesizability can jointly output predicted synthesis routes for generated molecules. However, there has been minimal attention in addressing the ease of synthesis and with flexibility to incorporate desired reaction constraints. On the other hand, virtual screening searches for commercially available compounds, but imposes challenges when scaling to ultra-large (billion-size and beyond) chemical spaces. Here, we propose a generative design framework that unifies synthesis-constrained molecular design and ultra-large-scale virtual screening through steerable and granular synthesizability control. Generated molecules satisfy arbitrary multi-parameter optimization objectives with predicted synthesis routes satisfying mix-and-match constraints: including or avoiding certain reactions, incorporating specific building blocks, and minimizing synthesis route length. In an end-to-end in-house campaign targeting BRD4, we designed molecules synthesizable with specific selected reactions and building blocks, synthesized all six selected compounds, and identified two micromolar binders. We further demonstrate that reaction control enables efficient navigation of ultra-large make-on-demand chemical spaces to identify property-optimal candidates. By applying our framework to Chemspace's Freedom 4.0 make-on-demand space (142 billion molecules), we generated ~320k molecules (0.00023% of the library) on a single consumer-grade GPU (with only 8 GB GPU memory) and identified a micromolar Wee1 binder amongst 60 synthesized candidates. The single unified framework thus enables generating novel synthesizable molecules and retrieving catalogue-ready candidates, offering a flexible solution to mitigating the synthesizability bottleneck.

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

Experimental Analysis of Neural Network-Based Image Classification on the CIFAR-10 Dataset

An experimental investigation of neural image classification on the CIFAR-10 benchmark is presented through fully connected and convolutional network formulations. The analysis emphasizes the complete learning pipeline: image vectorization, normalization, one-hot class encoding, supervised loss minimization, learning-rate selection, mini-batch training, convolutional feature extraction, max-pooling, and validation-based generalization assessment. A convolutional architecture with six convolutional layers and three max-pooling stages is evaluated for ten training epochs using a batch size of 128 and an Adam optimizer with a learning rate of 0.001. The validation accuracy reaches approximately 74.77%, while the validation loss begins to increase after the middle of training despite continued reduction in training loss. The resulting behavior illustrates the practical difference between representation learning and memorization, and it provides a compact experimental baseline for future studies on regularization, data augmentation, deeper architectures, and reproducible image-classification education.

13.
PLOS Computational Biology 2026-06-18

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

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

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

HalluJudge: A Reference-Free Hallucination Detection for Context Misalignment in Code Review Automation

arXiv:2601.19072v3 Announce Type: replace-cross Abstract: Large Language models (LLMs) have shown strong capabilities in code review automation, such as review comment generation, yet they suffer from hallucinations – where the generated review comments are ungrounded in the actual code – poses a significant challenge to the adoption of LLMs in code review workflows. To address this, we explore effective and scalable methods for a hallucination detection in LLM-generated code review comments without the reference. In this work, we design HalluJudge that aims to assess the grounding of generated review comments based on the context alignment. HalluJudge includes four key strategies ranging from direct assessment to structured multi-branch reasoning (e.g., Tree-of-Thoughts). We conduct a comprehensive evaluation of these assessment strategies across Atlassian's enterprise-scale software projects to examine the effectiveness and cost-efficiency of HalluJudge. Furthermore, we analyze the alignment between HalluJudge's judgment and developer preference of the actual LLM-generated code review comments in the real-world production. Our results show that the hallucination assessment in HalluJudge is cost-effective with an F1 score of 0.85 and an average cost of $0.009. On average, 67% of the HalluJudge assessments are aligned with the developer preference of the actual LLM-generated review comments in the online production. Our results suggest that HalluJudge can serve as a practical safeguard to reduce developers' exposure to hallucinated comments, fostering trust in AI-assisted code reviews.

15.
arXiv (CS.CL) 2026-06-18

Trade-offs in Medical LLM Adaptation: An Empirical Study in French QA

The development of large language models (LLMs) has led to an increased focus on their adaptation to specialized domains and languages, yet the effectiveness of domain adaptation strategies remains unclear. We present a study of medical domain adaptation using French medical question-answering (QA) as a case study. We compare continual pretraining (CPT), supervised fine-tuning (SFT), and their combination across three model families, multiple sizes, and three initialization types, explicitly disentangling adaptation effects from base model choice. We evaluate both multiple-choice (MCQA) and open-ended QA (OEQA) under greedy and constrained decoding using automatic metrics and LLM-as-a-Judge evaluation. For MCQA, CPT+SFT most often achieves the best scores, but gains over SFT are small and frequently not statistically significant, making SFT a strong and cost-effective default. For OEQA, CPT consistently improves overlap-based metrics, while SFT often degrades generation quality; instruction tuning and CPT+SFT are preferred by LLM-based evaluation. Cross-lingual experiments further show effective transfer from French adaptation to English benchmarks. Overall, we provide practical guidelines for selecting adaptation strategies under computational constraints.

16.
arXiv (math.PR) 2026-06-18

A simple approach to the L{\o}kka-Zervos dichotomy for absolutely continuous dividend strategies

arXiv:2604.13302v3 Announce Type: replace-cross Abstract: We revisit the optimization problem solved in L{\o}kka & Zervos (2008), i.e., the maximization of dividends, in a Brownian risk model, with the possibility (not the obligation) of making capital injections. Following the approach introduced in Alvarez & Shepp (1998), Renaud & Simard (2021), Renaud et al. (2023), we consider instead absolutely continuous (AC) dividend strategies with an affine bound on the payment rates, while singular capital injections are still allowed. In addition, we incorporate a parameter for the cost of ruin or, said differently, a penalty at ruin in the performance function. We show that the solution is a so-called L{\o}kka-Zervos dichotomy: the surplus is never ruined by making bail-out payments, or no capital is injected and bankruptcy can occur; in either case, dividends are paid at full rate when the surplus is above a threshold. Our framework allows us to provide explicit conditions to express the dichotomy, either using the cost of capital injections or the cost of ruin as a criterion, which also exposes the underlying structure of the solution. In particular, for some values of the parameters, we show that it is optimal to liquidate. Moreover, we perform a numerical analysis highlighting the range of values generated under this AC affine-bound structure.

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

End-to-End Machine Learning for Depressive State Classification via EEG and fNIRS

arXiv:2606.11555v1 Announce Type: cross Abstract: The escalating demand for mental healthcare, driven by rising societal stress, highlights the limitations of traditional psychiatric diagnostics. Conventional methods - relying primarily on clinical interviews and patient self-reports - are inherently vulnerable to subjective bias and the varying empirical judgment of practitioners. To address the need for quantitative evaluation, biological signal-based detection, including electroencephalography (EEG) and functional near-infrared spectroscopy (fNIRS), has emerged as a promising objective alternative. Such technology is particularly vital for identifying latent depressive states that may be unrecognized by the subjects themselves. Furthermore, in aging populations, the high comorbidity between depression and dementia necessitates early differentiation to prevent mutual symptom exacerbation and maintain Quality of Life (QoL). This pilot study of eleven healthy students establishes a framework for biological signal-based depression detection, serving as a foundational step toward automated, objective diagnostic tools for clinical use.

18.
medRxiv (Medicine) 2026-06-22

Early-life nutritional environment is associated with late-life cognition in the Health and Retirement Study, a pellagra epidemic natural experiment

Early-life exposures are important to several late-life health outcomes. We sought to study the effect of an in utero nutritional environment and its interaction with Alzheimer's disease (AD) genetic risk on late-life cognitive function. We used a natural experiment created by the pellagra epidemic, a nutritional disease caused by a vitamin B3 deficiency, to evaluate the association between in utero pellagra epidemic exposure and late-life cognitive function in the Health and Retirement Study (N = 18,285). We also evaluated whether the in utero exposure could modify the AD polygenic score's (PGS) effect on cognition. In utero pellagra epidemic exposure was significantly associated with cognition ({beta} = -0.025). However, these effects were not isolated to the prenatal period as exposure during childhood periods also had an effect. The interaction between the in utero exposure and the AD PGS was significant, where the genetic effect on cognition was amplified with increasing (progressively worse) in utero exposure levels. These associations imply that the early-life nutritional environment affects late-life cognitive function and that these effects can modify genetic risk.

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

RouteJudge: An Open Platform for Reproducible and Preference-Aware LLM Routing

arXiv:2606.18774v1 Announce Type: new Abstract: We present RouteJudge, an online pairwise preference evaluation framework for LLM routing systems, with a public platform available at https://routejudge.cn. Different from model-level response evaluation, RouteJudge focuses on router-level decision quality. For each user query, multiple routing strategies independently recommend candidate models under the same model pool and budget constraints. The selected model responses are then presented to users through anonymous pairwise comparisons, and the resulting user preferences are attributed back to the routing strategies behind the compared responses. Each evaluation record stores the query, routing decisions, model responses, preference labels, cost, latency, and task metadata, enabling preference-aware, cost-aware, and task-conditioned analysis of LLM routers. To support the continuous expansion of routing methods in RouteJudge, we further release ORBIT (Optimal Routing and Budgeted Inference Toolbox), a modular and extensible toolbox that standardizes the end-to-end workflow of LLM routing. ORBIT provides unified interfaces for benchmark loading, query representation, router implementation, budget-aware evaluation, and method comparison, allowing researchers to develop and evaluate routing algorithms under consistent protocols. It also serves as the submission and integration layer for RouteJudge: researchers can implement routing methods within ORBIT, validate them on existing routing benchmarks, and submit compatible routers for online preference-based evaluation. The code of ORBIT is available at https://github.com/AIGNLAI/LAMDA-ORBIT.

20.
arXiv (CS.CL) 2026-06-12

Small LLMs for Biomedical Claim Verification: Cost-Effective Fine-Tuning, Structural Dataset Shortcuts, and Cross-Domain Generalization

作者:

Large Language Models such as GPT-4o and GPT-5 achieve strong zero-shot performance on biomedical claim verification, but cost and opacity limit scalable use. We fine-tune three small LLMs: Phi-3-mini (3.8B), Qwen2.5-3B, and Mistral-7B, via QLoRA on SciFact and HealthVer, providing the first study of QLoRA models against GPT-4o and fine-tuned BioLinkBERT encoders. Mistral-7B QLoRA surpasses both GPT-4o and GPT-5 (up to 12% F1 gain) at a fractional cost using just 1,008 training examples. We conduct extensive in-domain and cross-domain evaluation: models trained on SciFact tested on HealthVer and vice versa, at matched sizes to isolate dataset structure from data quantity. We identify a previously unreported structural artifact in SciFact that inflates in-domain scores, and show through bidirectional out-of-domain evaluation that training on structurally sound data enables robust cross-domain transfer. We plan to release all code and adapter checkpoints.

21.
arXiv (CS.LG) 2026-06-18

Diffusion-Proof: Recipe for Formal Theorem Proving Beyond Auto-Regressive Generation

arXiv:2606.19315v1 Announce Type: new Abstract: Enhancing the formal math reasoning capabilities of Large Language Models (LLMs) has become a key focus in both mathematical and computer science communities in recent years. While significant progress has been made in using state-of-the-art Auto-Regressive (AR) LLMs for formal theorem proving, these models suffer from inherent limitations. Their next-token prediction generation methods may yield suboptimal performance due to the challenges of long-range coherence and the compounding of errors over long sequences. Recent advancements in diffusion LLMs (dLLMs), which generate text through iterative denoising of a multi-token block, offer a promising alternative. However, the application of dLLMs to formal mathematics, where maintaining long-range coherence is critical, remains largely understudied. To address the challenges above, we propose **Diffusion-Proof**, to the best of our knowledge, the first framework to train and apply dLLMs for formal theorem proving. Our frameworks contain training and inference methods for two models. The first one is *dLLM-Prover-7B*, which performs whole-proof writing with long-range coherent tactic usage. The second one is *dLLM-Corrector-7B*, which is a novel large block diffusion-based correction model. It leverages the in-filling capabilities of dLLMs to perform local proof correction using bi-directional information. Extensive experiments demonstrate that **Diffusion-Proof** relatively significantly outperforms the AR LLM baseline trained under the same dataset. **Diffusion-Proof** achieves an absolute improvement of **1.61%** on ProofNet-Test and **6.14%** on MiniF2F-Test benchmarks compare to the baseline. Notably, **Diffusion-Proof** successfully resolves one IMO problem that more advanced thinking model DeepSeek-Prover-V2-7B could not solve, showcasing the unique advantage of dLLMs in formal theorem proving.

22.
arXiv (quant-ph) 2026-06-12

Beyond-Third-Order Quantum Coherence in Two-Dimensional Spectroscopy via Order-Selective Isolation

arXiv:2606.12794v1 Announce Type: new Abstract: A central challenge in nonlinear spectroscopy is the order-selective readout of weak higher-order responses that spectrally overlap with dominant lower-order signals. This bottleneck is particularly severe in two-dimensional (2D) spectroscopy, where extending conventional phase-cycling schemes to higher orders rapidly increases measurement and analysis complexity. Here we introduce a computation-assisted strategy that combines rotating-frame acquisition with a frame-shift tracking algorithm to separate signals by their frame-dependent spectral shifts. In a rubidium vapor experiment, we use this approach to isolate a 7th-order nonlinear contribution from coexisting 3rd-order components, enabling direct access to higher-order quantum-coherence dynamics without sacrificing operation at comparatively high pulse intensities. The method is broadly compatible with multidimensional spectroscopy platforms and provides a practical route to probing many-body and collective ultrafast dynamics beyond third order.

23.
arXiv (CS.CV) 2026-06-17

SPHINX: First Explain, Then Explore

Generating adversarial driving scenarios is critical for evaluating and improving autonomous vehicle decision-making systems in simulation. Recent approaches, such as ChatScene and LLM-Attacker, rely primarily on the prior knowledge of Large Language Models and Vision-Language Models to generate driving scenarios procedurally. We argue that adversarial scenes should be generated based on the failure diagnosis (e.g., indecisiveness, multi-frame inconsistency) of the driving policy to specifically address the policy's weaknesses instead of relying on prior assumptions. In this paper, we propose SPHINX, a closed-loop framework for adversarial scenario synthesis guided by a simple principle: first explain, then explore. Beyond blindly exploring the scenario space, SPHINX leverages explainable artificial intelligence methods to analyze the policy, identifying key visual concepts and their influence on policy outputs, and the uncertainty of the decisions. Given the interpretable evidence extracted from the policy's own decision process, we use a vision language model to rationalize and criticize failure modes of the current policy. These critics are then used to generate targeted adversarial scenarios for policy retraining and improvement. We demonstrate that SPHINX can highlight an interpretable account of policy failures while other adversarial scene generation cannot. Across the evaluated benchmarks and test suites, SPHINX can be applied to diverse state-of-the-art autonomous vehicle architectures and yields consistent robustness improvements over existing scenario-generation methods.

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

Guiding Federated Graph Recommendation with LLM-encoded knowledge

arXiv:2606.15277v1 Announce Type: cross Abstract: Graph-based recommender systems are highly effective at extracting collaborative signals from user–item interactions, and federated learning (FL) allows these models to be trained while preserving user privacy. However, aggregating graph representations across distributed, non-IID clients remains a challenge; structural embeddings learned locally often misalign, and naive averaging fails to capture meaningful cross-client relationships. Most existing federated graph methods rely exclusively on structural aggregation, neglecting the rich, global semantic context available in large language models (LLMs). In this paper, we propose a novel framework that uses LLM-encoded knowledge to guide federated graph recommendation. Specifically, clients learn structural representations from local graphs while simultaneously summarizing their typical interaction patterns into compact semantic vectors via a frozen LLM. The central server then uses these LLM-encoded semantic signals to discover related preference patterns across clients, guiding the selective aggregation of their structural representations. This enables semantically informed cross-client collaboration without exposing raw data. Extensive experiments on standard benchmarks show that guiding structural alignment with LLM-encoded knowledge consistently improves recommendation accuracy over existing federated graph baselines.

25.
arXiv (CS.LG) 2026-06-19

Sparsity, Superposition, and Forgetting: A Mechanistic Study of Representation Retention in Continual Learning

arXiv:2606.20431v1 Announce Type: new Abstract: Continual learning (CL) systems often forget previously acquired knowledge, yet the mechanisms driving forgetting remain hard to isolate in practice because real datasets entangle many factors. We present a controlled, toy-world framework that makes these mechanisms observable and testable. Using a synthetic generator-separator pipeline, we define ground-truth latent features, build tasks with tunable sparsity and overlap, and introduce measurable quantities for representation strength and superposition (directional overlap among features). We then study retention dynamics-the temporal change of representation strength by fitting sparse dynamical relations (via SINDy) between retention, superposition, and exposure history. A complementary task-level analysis based on effective rank characterizes how representational capacity is allocated across tasks. Our controlled experiments yield three takeaways. (1) Superposition tends to increase over time with transient dips at task boundaries, suggesting boundary-specific interference rather than steady drift. (2) Higher feature sparsity induces more superposition yet does not inevitably cause forgetting; when representations remain strong, forgetting can be reduced despite overlap. (3) Task-level effective rank grows with sparsity, indicating broader capacity usage under sparse regimes. Together, these results nuance the common intuition that more superposition leads to more forgetting by showing that overlap interacts with representation strength and capacity allocation. Our toy analysis provides falsifiable hypotheses and diagnostic tools for CL.