Academic Intelligence · Curated Daily

探索全球前沿学术脉络

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

01.
arXiv (CS.CL) 2026-06-11

LibriConvo: Simulating Conversations from Read Literature for ASR and Diarization

We introduce LibriConvo, a synthetic conversational speech corpus for speaker diarization and automatic speech recognition (ASR), built by instantiating the previously proposed Speaker-Aware Simulated Conversation (SASC) framework in a dataset and benchmarking setting. The main contribution of this paper is a corpus construction pipeline and benchmark derived from that framework. To make the data more suitable for downstream ASR and diarization, conversational timing statistics are estimated from English CallHome using external voice activity detection, long pauses are compressed, LibriTTS utterances are grouped by book to improve local semantic continuity, and room impulse responses are selected with a spatial-plausibility heuristic. The resulting corpus contains 240.1 hours of audio across 1,496 dialogues involving 830 speakers, partitioned into speaker-disjoint train, validation, and test splits. We report baseline results for both diarization and ASR. On the test split, Sortformer outperforms the pyannote pipeline in diarization (11.1\% vs.~24.4\% DER). For ASR, a Fast Conformer-CTC XLarge model fine-tuned with Serialized Output Training achieves 7.29\% WER and 6.97\% cpWER, outperforming zero-shot Whisper-large-v3. These results position LibriConvo as a practical benchmark for studying synthetic conversational speech and for evaluating multi-speaker speech processing systems.

02.
arXiv (CS.CV) 2026-06-11

SpikeTAD: Spiking Neural Networks for End-to-End Temporal Action Detection

Video understanding is a crucial part of computer vision, with numerous application scenarios. With the increasing popularity of mobile devices, an increasing number of efforts are trying to deploy video understanding models on them. However, existing video understanding models are difficult to deploy due to their large size and prohibitive power consumption. Spiking Neural Networks (SNNs) have shown bioplausibility and low power advantages over Artificial Neural Networks (ANNs), especially on neuromorphic chips which are regarded as essential components of future mobile devices. However, excessively long conversion time-steps and severe performance degradation problems limit their application. To solve the problems above, we explore the application of SNNs on temporal action detection (TAD), which is an important task in video understanding, and propose the first SNN-based end-to-end TAD architecture coined as SpikeTAD. While maintaining extremely low power consumption, SpikeTAD achieves an average mAP of 67.2% in THUMOS14 and 37.42% in ActivityNet-1.3, demonstrating the feasibility of a low-power TAD model. Our code is available at https://github.com/MCG-NJU/SpikeTAD.

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

Generalised Eigenvalue Geometry of Semantic Adversarial Attacks

arXiv:2606.19212v1 Announce Type: cross Abstract: Recent empirical work shows that semantically equivalent paraphrases can fool financial sentiment classifiers: although a paraphrase remains close to the original under a strong reference embedding, it may shift the target model's representation enough to change the predicted class. Existing robustness theory either assumes a single-model threat model or focuses mainly on empirical attack algorithms. We develop a continuous local model of semantic paraphrase perturbations that captures this two-model structure. We show that the worst-case local displacement of the target representation, subject to a proxy-model budget, is governed by the largest generalised eigenvalue of a matrix pencil $(A,B)$ constructed from the Jacobians of the two embedding maps. The resulting attackability index $\lambda^*(x)$ is intrinsic to the local paraphrase geometry and the chosen embedders, yields a closed-form prediction-flip condition for affine readouts, and supports conservative population and finite-sample attackability certificates. For uniform control over classes of affine readouts, we derive a distribution-free VC bound for binary attackability indicators and a scale-sensitive margin bound based on an attackability-adjusted margin that subtracts a local geometric penalty from the standard classifier margin. We also connect the continuous theory to discrete paraphrase search, identify an asymmetry between successful and unsuccessful finite searches, and give a covering condition under which the discrete and continuous settings agree. Finally, we propose an empirical verification framework using soft-token relaxations and generated paraphrase sets to assess the local eigenvalue geometry, prediction-flip condition, and finite-search approximation on a deployed financial-text classifier.

04.
medRxiv (Medicine) 2026-06-17

Trends in Suicide Mortality by Method among US Individuals aged 10-24 Years from 1999 to 2024

Background: Suicide is the second leading cause of death in US adolescents aged 10-24. Method use strongly influences lethality and design of prevention strategies, but recent trends remain unclear. We therefore aimed to investigate trends in suicide mortality rates by method, age group, and sex. Methods: This cross-sectional study used suicide mortality data from the National Center for Health Statistics for a quarter-century period, between 1999 and 2024. All individuals aged 10-24 years at the time of death, with suicide as the underlying cause, were included. We estimated suicide mortality rates (i.e., the number of suicide deaths per 100,000 people) and annual percent change by method (firearm, asphyxiation, poisoning, other), age group (10-14, 15-19, 20-24), and sex. Changing trend time points were determined using Joinpoint regression models Results: From 1999 to 2024, 159,241 suicide deaths occurred among individuals aged 10-24. While suicide rates declined across all age groups between 2017 and 2024, the male-to-female gap narrowed by 18.9%. Among 10-14-year-olds, declining rates among males masked a consistent increase in female suicide rates since 2011. Although asphyxiation-related suicides decreased across all groups since 2018, firearm suicide rates increased for females in the 10-14 and 20-24 age groups. Albeit not as common as firearms or asphyxiation, poisoning suicide rates increased in the 15-19 and 20-24 age groups. Since 1999, suicide rates by other less common methods (e.g., jumping) showed significant increases, for both sexes, especially among individuals aged 20-24. Suicide rates were consistently highest in the 20-24 age group across all study years. Conclusion: The decrease in suicide mortality rates among individuals aged 10-24 was largely driven by declines in males and reductions in asphyxiation-related suicides. However, increasing female suicide rates in the 10-14 age group, as well as increasing rates of death by less common means, warrant close attention. While suicide prevention efforts like structural interventions and means restriction have shown effectiveness among male adolescents, priority should now be given to adapting these approaches for female adolescents, particularly those aged 10-14.

05.
arXiv (quant-ph) 2026-06-12

Information gain and measurement disturbance for quantum agents

arXiv:2402.08060v3 Announce Type: replace Abstract: The traditional formalism of quantum measurement (hereafter ``TQM'') describes processes where some properties of quantum states are extracted and stored as classical information. While TQM is a natural and appropriate description of how humans interact with quantum systems, it is silent on the question of how a more general, quantum, agent would do so. How do we describe the observation of a system by an observer with the ability to store not only classical information but quantum states in its memory? In this paper, we extend the idea of measurement to a more general class of sensors for quantum agents which interact with a system in such a way that the agent's memory stores information (classical or quantum) about the system under study. For appropriate sensory interactions, the quantum agent may ``learn'' more about the system than would be possible under any set of classical measurements – but as we show, this comes at the cost of additional measurement disturbance. We experimentally demonstrate such a system and characterize the tradeoffs by considering the channel capacity required to erase the effect of a measurement.

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

The Accountability Paradox: How Platform API Restrictions Undermine AI Transparency Mandates

arXiv:2505.11577v5 Announce Type: replace-cross Abstract: Recent application programming interface (API) restrictions on major social media platforms challenge compliance with the EU Digital Services Act [20], which mandates data access for algorithmic transparency. We develop a structured audit framework to assess the growing misalignment between regulatory requirements and platform implementations. Our comparative analysis of X/Twitter, Reddit, TikTok, and Meta identifies critical ``audit blind-spots'' where platform content moderation and algorithmic amplification remain inaccessible to independent verification. Our findings reveal an ``accountability paradox'': as platforms increasingly rely on AI systems, they simultaneously restrict the capacity for independent oversight. We propose targeted policy interventions aligned with the AI Risk Management Framework of the National Institute of Standards and Technology [80], emphasizing federated access models and enhanced regulatory enforcement.

07.
arXiv (CS.AI) 2026-06-11

Are Frontier LLMs Ready for Cybersecurity? Evidence for Vertical Foundation Models from Dual-Mode Vulnerability Benchmarks

arXiv:2605.23243v2 Announce Type: replace-cross Abstract: We evaluate whether frontier LLMs are ready for cybersecurity through a dual-mode benchmark: white-box function-level vulnerability detection (VulnLLM-R, across C/Java/Python) and black-box web application security testing (five production-style applications with 118 ground-truth vulnerabilities across 20+ CWE families, which we will open-source). We test six frontier models (GPT-5.4, Codex~5.3, Claude Opus~4.6, Sonnet~4.6, Gemini~3.1~Pro and Gemini~3~Flash) and two domain-specialized models across four testing paradigms. Our findings are sobering: (1)~every frontier model produces 10-50% false positive rates in white-box detection, systematically over-predicting vulnerabilities; (2)~in black-box testing, frontier models achieve only 4-8% ground-truth coverage, improving to just 10-19% even with external security tools (Playwright MCP, Burp Suite MCP); (3)~structured penetration-testing methodology encoded in domain-specialized agents raises per-family detection above 50%, demonstrating that methodology, not scale, is the primary lever; and (4)~a domain-specialized defense model achieves the highest precision (0.904) and lowest false positive rate (9.7%) among all models, on a single GPU. We identify the absence of structured security testing traces end-to-end request/response sequences, failure-heavy data, and multi-step attack chains as the fundamental training data bottleneck, and propose self-play security testing as a data generation strategy. Our results make the case for vertical foundation models purpose-built for cybersecurity.

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

Unsupervised Learning for Missing Modalities in Multimodal Learning

arXiv:2606.15743v1 Announce Type: new Abstract: This paper addresses the missing-modality challenge in multi-modal learning by introducing Unsupervised Learning for Missing Modalities in Multi-Modal Learning (UL4M4), a flexible framework that imputes missing feature embeddings in a task-independent manner before supervised prediction. We propose modality-specific normalization and a novel partial-modality distance metric to enable fair clustering of incomplete observations, capturing cross-modal structures while preserving scale-invariance across varying dimensionalities and modality counts. Cluster centers from this unsupervised stage guide an iterative greedy imputation process for any missing modalities during training or inference, supporting arbitrary numbers of modalities and arbitrary missing patterns per sample. The imputation module is lightweight, uses frozen encoders, and decouples from the downstream task, allowing easy integration with any fusion/prediction architecture. Extensive experiments under diverse and highly incomplete regimes demonstrate UL4M4's robustness, achieving, to the best of our knowledge, the first consistent F1-Micro scores above 0.7 on challenging missing configurations even when more than 50\% of modality slots are missing. Results are also stable across cluster sizes and significantly outperform state-of-the-art baselines. Code is available here: https://github.com/h-ismkhan/Multimodal-Learning-with-Missing-Modalities-via-Unsupervised-Learning.

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

Optimal Spatio-Temporal Decoupling for Bayesian Conformal Prediction

arXiv:2605.00432v2 Announce Type: replace Abstract: Online conformal prediction must balance fast adaptation to distribution shift against stable coverage: feedback-driven methods react quickly but become volatile, while strongly discounted Bayesian methods lag and inflate intervals at tight coverage. We introduce State-Adaptive Bayesian Conformal Prediction (SA-BCP), which forms the predictive quantile as a gated convex combination of long-term temporal inertia and local spatial evidence from a kernel density estimate, controlled by a single interpretable evidence threshold $K$. We establish three results: (i) asymptotic marginal validity of the resulting intervals; (ii) a closed-form expression for the MSE-optimal threshold, $K^*_{\mathrm{MSE}}=\alpha(1-\alpha)/M^{\mathcal{T}}$, trading the coverage-indicator (Bernoulli) variance against the temporal structural bias $M^{\mathcal{T}}$; and (iii) a rolling-origin procedure for selecting $K$ online – consistent under stationarity, with $O(\sqrt{T\log N})$ regret against the best fixed $K$ and, for a segmented variant, a sublinear dynamic-regret bound under bounded drift. Across four financial-volatility and weather datasets, three target coverage levels, and eight baselines (including the strongest recent conditional-quantile methods, SPCI and KOWCPI), SA-BCP attains at-or-above-nominal coverage in most settings while producing substantially sharper intervals – up to roughly $3\times$ lower Winkler score than discounted Bayesian CP at the tightest coverage – and a coverage-matched audit confirms these efficiency gains are not an artifact of under-coverage. We disclose one principal limitation: a volatility-specialized conformal-GARCH competitor remains more efficient on its home volatility-base series, though it does not transfer across domains.

10.
arXiv (CS.CL) 2026-06-11

When Roleplaying, Do Models Believe What They Say?

Language models can state that "the Earth orbits the Sun" and, when role-playing Aristotle, assert the opposite. Recent work argues that persona adoption is fundamental to how language models operate, with models constantly selecting the most appropriate persona for a given context. Does such role-playing merely change the model's outputs, or does it also affect what the model internally represents as truthful? We study this question with linear truth probes, applying them to LLMs role-playing historical personas whose likely beliefs differ from modern consensus. For each persona, we compare false claims the persona would likely have endorsed (*era-believed*) with topic-matched false claims they would not have endorsed (*era-false*). Across prompting, in-context learning, and supervised fine-tuning, persona induction suppresses era-believed statements less than equally false alternatives, yet they remain classified as false overall. Role-play therefore shifts what these models say more than what they internally represent as true. We contrast this with models trained on harmful advice that exhibit Emergent Misalignment (EM). Across three model families (Qwen 2.5 14B, Qwen 3 8B, and Llama 3.3 70B), their false claims move substantially toward the true region of probe space, are defended under challenge roughly half the time versus about a sixth for role-play, and are used in downstream reasoning. Role-play and Emergent Misalignment thus are points on a spectrum of belief internalization, where role-play changes what a model says with little representational change, while Emergent Misalignment shifts the internal representation of false claims without fully marking them as true.

11.
arXiv (CS.CL) 2026-06-24

Mind the Heads: Topological Representation Alignment for Multimodal LLMs

Representation alignment has emerged as an effective approach to improve Multimodal Large Language Models (MLLMs) by regularizing their internal representations toward those of an external vision encoder. However, existing methods typically align a fixed layer of the language backbone, overlooking the fine-grained structure of Transformer models. In this work, we propose Head-Wise Representation Alignment (HeRA), a method that enforces cross-modal alignment at the level of individual attention heads. Our approach is grounded in the Platonic Representation Hypothesis, focusing on preserving the topological structure of representations (i.e., their local neighborhood relationships) across modalities. Following the Mutual K-Nearest Neighbor (MKNN) alignment metric, we introduce a contrastive objective that acts as a differentiable proxy for matching local structures. HeRA applies this objective during multimodal training to specific attention heads in the LLM, selected by their alignment score according to the MKNN metric. Counterintuitively, we find that aligning the least aligned heads yields the largest gains. Extensive evaluations across multiple MLLMs and 18 benchmarks demonstrate that HeRA consistently improves performance on challenging vision-centric tasks and serves as an effective regularizer against visual hallucinations by naturally curbing the over-reliance on linguistic priors. Our code is publicly released.

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

Exact Federated Continual Unlearning for Ridge Heads on Frozen Foundation Models

arXiv:2603.12977v3 Announce Type: replace Abstract: Foundation models are commonly deployed as frozen feature extractors with a small trainable head to adapt to private, user-generated data in federated settings. The ``right to be forgotten'' requires removing the influence of specific samples or users from the trained model on demand. Existing federated unlearning methods target general deep models and rely on approximate reconstruction or selective retraining, making exactness costly or elusive. We study this problem in a practically relevant but under-explored regime: a frozen foundation model with a ridge-regression head. The exact optimum depends on the data only through two additive sufficient statistics, which we turn into a communication protocol supporting an arbitrary stream of add and delete requests via fixed-size messages. The server maintains a head that is, in exact arithmetic, pointwise identical to centralized retraining after every request. We provide deterministic retrain-equivalence guarantees, order and partition invariance, two server-side variants, and a Bayesian certificate of zero KL divergence. Experiments on four benchmarks confirm the guarantees: both variants match centralized ridge retraining to within $10^{-9}$ relative Frobenius error and complete each request at orders-of-magnitude lower cost than federated retraining baselines.

13.
arXiv (CS.CV) 2026-06-16

Sub-Semantic Image Segmentation

Images can be segmented based on visual cues (i.e., texture segmentation) or into objects (i.e., semantic segmentation). We propose a new category of sub-semantic image segmentation that blurs the line between the two. In sub-semantic image segmentation, language is not used to name whole objects. Instead, it is used to partition an image into stable appearance patterns that can be described by language. To do that, we couple a general-purpose vision-language model to SAM 3, a promptable segmentation backbone whose native text pathway can ground rich descriptions into masks. Simple coupling fails for a number of reasons that we identify in the paper, and we overcome them by introducing DETECTURE that resolves three concrete failure modes – language leakage between texture regions, prompt competition inside the segmentation backbone, and semantic distortion at the language-to-mask interface. Since there is no dataset of sub-semantic image segmentation, we introduce one, termed TextureADE. The new dataset is derived from the ADE20K dataset using a system we designed. We compare DETECTURE to a number of baselines and find that it achieves the strongest performance on several datasets using different metrics. Code is available at https://github.com/Scientific-Computing-Lab/TextureDetecture.

14.
arXiv (CS.LG) 2026-06-11

Geometric bias in eigenspace perturbation under random heterogeneous noise

arXiv:2606.11263v1 Announce Type: cross Abstract: Spectral methods rely fundamentally on the stability of principal eigenspaces under random perturbations. Classically, this stability is quantified by the Davis-Kahan and Wedin theorems, which bound the eigenspace error using the operator norm of the noise and the relevant spectral gaps. While these worst-case bounds are sharp for arbitrary deterministic perturbations, they can be wasteful in the low-rank signal-plus-random-noise setting, as they fail to capture the fine-grained interaction between the signal geometry and the noise distribution. In this paper, we study the spectral perturbation of signal-plus-noise matrices corrupted by sparse, random noise with an arbitrary, inhomogeneous variance profile. We demonstrate that under heterogeneous noise variances, the empirical eigenvectors suffer a systematic, deterministic geometric bias that is entirely invisible to classical perturbation bounds. By leveraging the Quadratic Vector Equation (QVE) and establishing fine-grained isotropic local laws, we derive near-optimal, non-asymptotic perturbation bounds for the leading eigenspaces in the operator and $2\to\infty$ norms. The bounds separate the usual signal-to-noise contribution, stochastic fluctuations, and structured geometric bias terms determined by the alignment between the signal eigenspaces and the row-wise variance profile.

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

From Reasoning Traces to Reusable Modules: Understanding Compositional Generalization in Language Model Reasoning

arXiv:2606.18089v1 Announce Type: new Abstract: Post-training pipelines that combine supervised fine-tuning (SFT) with reinforcement learning (RL) have emerged as the key recipe for transforming large language models (LLMs) into robust reasoners. We argue that this combined success is driven by compositional generalization, which we formalize through a hierarchical latent selection model. In this framework, reasoning traces are generated by a cascade of discrete latent selection variables corresponding to reusable atomic modules, including both skills (local operations) and routing mechanisms (how intermediate information is selected, reused, and composed). Within this model, we theoretically show that SFT and RL play asymmetric, complementary roles: SFT supplies the raw module materials in compositional traces, and RL decomposes those traces to identify the latent atomic modules and enable compositional generalization. We design controlled experiments to validate this theory. Our results demonstrate that RL can extract atomic modules from compound traces supplied by SFT and recombine them to solve new configurations. Moreover, we find that training on compound traces yields stronger generalization than training on isolated atomic modules. Finally, we investigate the relationship between SFT and RL data and identify an effective protocol in which SFT ensures coverage of all atomic modules through compositional traces, while RL focuses on novel compositions outside the SFT support to drive exploration.

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

(Human) Attention Is (Still) All You Need: Human oversight makes AI-assisted social science reliable

arXiv:2606.12848v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly used for tasks once reserved for trained researchers, including hypothesis generation, specification choice, and drafting conclusions. We argue that the reliability of AI-assisted research depends not only on model capability, but also on how cognitive labour is structured between humans and machines. We study this problem through Human-in-the-Loop Economic Research (HLER), a decision architecture based on pre-commitment, decision sequencing, accountability, and attention allocation. In a pre-specified 2*4 factorial experiment with 280 complete research runs across four datasets, an unconstrained multi-agent baseline produced critical failures in 72% of runs. Using the same underlying model, the same agent decomposition, and identical prompts for the shared reasoning agents, HLER reduced the failure rate to 16% by imposing three architectural commitments: LLMs reason but do not execute data work, data and estimation are handled deterministically, and three human decision gates bind the workflow. Fisher's exact test rejects equality of failure rates at p

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

ConSA: Controllable Sparsity in Hybrid Attention via Learnable Allocation

Hybrid architectures combining full attention (FA) and sliding-window attention (SWA) are a promising paradigm for efficient LLM inference. However, existing methods typically rely on hand-crafted rules or simple post-hoc heuristics for FA/SWA allocation and offer limited analysis of the attention behaviors underlying these designs. We propose Controllable Sparsity in Hybrid Attention (ConSA), a framework that learns optimal FA/SWA assignment under a user-specified sparsity target. ConSA employs L0 regularization to learn binary masks selecting between FA and SWA for each attention unit, while an augmented Lagrangian constraint enforces the target sparsity at either layer or KV-head granularity. We evaluate ConSA on two LLMs at the 0.6B and 1.7B scales. Learned allocations consistently outperform rule-based baselines, with KV-head-wise allocation yielding clear gains over layer-wise allocation. The learned patterns place SWA in the bottom layers and concentrate FA into contiguous middle-layer blocks, diverging from evenly interleaved patterns in rule-based methods. This structure persists across model scales, sparsity levels, and allocation granularities, revealing a fine-grained spectrum of intrinsic attention behaviors that underlies the learned allocation.

18.
arXiv (CS.AI) 2026-06-16

Minimal Oversight: Uncertainty-Aware Governance for Delegated AI Systems

arXiv:2606.15563v1 Announce Type: new Abstract: AI systems increasingly delegate decisions to specialized models, evaluators, tools, and supervisory controllers. The central AI problem is no longer only model accuracy, but uncertainty-aware governance: how much autonomy to grant, which evidence should calibrate trust, what performance ceiling a delegated AI system can sustain, and when human intervention becomes necessary. We propose the Minimum Sufficient Oversight Principle (MSO), a variational principle for principled autonomy delegation: minimize governance burden on the Fisher information manifold subject to a delivery constraint. The resulting Euler-Lagrange solution yields a water-filling allocation of governed delegation across the task space. Building on a revealed-action governed delegation channel model, we prove a capacity theorem for stationary symbolwise review policies, derive a local first-order approximation relating workflow complexity to quality degradation, and give a drift-dominated autonomy-time scaling law linking intervention timing to effective capacity, complexity, and drift. Within this framework, masking appears as a structural AI-governance pathology: corrected performance can hide the competence signal needed to calibrate trust. Synthetic simulations and a semi-real reconstructed workflow support design prescriptions including upstream-first correction, sensitivity-based intervention, and explicit feasibility checks before autonomy is expanded. The result is a computable framework for uncertainty, planning, and oversight in delegated AI systems. A companion Python package is available at https://github.com/crbazevedo/delegation-lab.

19.
arXiv (CS.CV) 2026-06-15

Value-order Decomposition for Generalist Anomaly Detection

Industrial anomaly detection suffers from limited data, making cross-domain generalization particularly challenging. Generalist Anomaly Detection (GAD) aims to train a unified model on a source domain that can effectively detect anomalies in unseen target domains. In the initial semantic feature space, strong entanglement between anomalies and object categories or defect types hinders effective generalization across domains. Recent works address this issue by projecting features into a residual space; however, such methods primarily increase cross-domain overlap for normal features, while anomalous features remain specific to object categories, defect types and data domains, leading to poor alignment and generalization. To address this limitation, we propose Value-order Decomposition (VOD), a simple yet effective technique that bridges three types of generalization gaps across object categories, defect types (including real and synthetic defects), and data domains. VOD disentangles and suppresses object-category-, defect-type-, and domain-specific information, promoting alignment within normal and abnormal samples while preserving their separability, thereby enabling robust generalization across the three gaps. Leveraging the strong alignment between real and synthetic defects within the same object, we perform anomaly detection using only normal and synthetic-abnormal reference, and effectively generalize to unseen real defect types. Experiments on diverse industrial and medical benchmarks demonstrate that our method, using a simple cut-and-paste anomaly simulation strategy, achieves strong generalization across the three gaps.

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

Shopping Reasoning Bench: An Expert-Authored Benchmark for Multi-Turn Conversational Shopping Assistants

Conversational shopping assistants now serve hundreds of millions of customers, yet no existing benchmark jointly evaluates the open-ended multi-turn reasoning, domain expertise, and criterion-level quality that real shopping conversations demand. Shopping reasoning is unique among language model applications. Unlike factual question answering or verifiable code generation, it requires balancing subjective preferences, budget constraints, and cross-product trade-offs across multi-turn dialogue, capabilities absent from previous e-commerce and general-purpose benchmarks. We introduce the Shopping Reasoning Bench, an expert-authored benchmark of 525 missions (232 single-turn, 293 multi-turn) with 10863 importance-weighted binary rubrics authored by retail domain experts. These criteria are organized under a taxonomy of five reasoning categories and fifteen subcategories covering diverse demands such as preference refinement, trade-off analysis, and compatibility assessment. An evaluation of nine models across three families (GPT, Claude, Gemini) shows that pass rates reach only 57–77% overall. On multi-turn missions, all models score 13–29 points lower on optional above-and-beyond criteria than on required ones, and performance degrades 4–18 points as conversations progress. These gaps show that current models handle basic shopping assistance but fall short of expert-level advice, making Shopping Reasoning Bench a challenging testbed for future shopping assistant development.

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

Compatibility-Aware Dynamic Fine-Tuning for Large Language Models

Supervised Fine-Tuning (SFT) is the predominant paradigm for aligning large language models (LLMs), yet it suffers from optimization instability and limited generalization. Recent work attributes this issue to pathological gradient scaling and proposes Dynamic Fine-Tuning (DFT) to correct it at the token level. However, DFT assumes all demonstrations are equally suitable learning targets, an assumption violated by the strong heterogeneity of large-scale instruction data, where demonstration-policy mismatch induces high-variance updates at the sample level. We introduce Compatibility-Aware Dynamic Fine-Tuning (CADFT), a principled extension of DFT that controls sample-level optimization variance. CADFT derives a dynamic, policy-dependent compatibility signal from model likelihoods to modulate supervised updates, suppressing high-variance gradients from incompatible demonstrations. We further propose a delayed, low-frequency compatibility-guided rewriting strategy to transform persistently incompatible demonstrations into learnable targets. We show that CADFT can be interpreted as a variance-controlled estimator that generalizes token-level stabilization in DFT to the sample level. Extensive experiments demonstrate improved stability, generalization, and cold-start reinforcement learning initialization, while remaining fully supervised and independent of explicit reward modeling.

22.
arXiv (CS.AI) 2026-06-12

Structured vs. Unstructured Pruning: An Exponential Gap

arXiv:2603.02234v3 Announce Type: replace-cross Abstract: The Strong Lottery Ticket Hypothesis (SLTH) states that large, randomly initialized neural networks contain sparse subnetworks capable of approximating a target function at initialization without training, suggesting that pruning alone is sufficient. Pruning methods are typically classified as unstructured, where individual weights can be removed from the network, and structured, where parameters are removed according to specific patterns, as in neuron pruning. Existing theoretical results supporting the SLTH rely almost exclusively on unstructured pruning, showing that logarithmic overparameterization suffices to approximate simple target networks. In contrast, neuron pruning has received limited theoretical attention, despite its practical appeal for direct hardware speedups. In this work, we consider the problem of approximating a single bias-free ReLU neuron by pruning hidden units of a randomly initialized two-layer ReLU network, effectively isolating the intrinsic limitations of neuron pruning. We show that achieving an $\varepsilon$-approximation requires a starting network size of $\Omega(1/\varepsilon)$ for neuron pruning, whereas weight pruning succeeds with only $O(\log(1/\varepsilon))$ hidden units, revealing an exponential separation between the two approaches.

23.
arXiv (CS.AI) 2026-06-19

"**Important** You should give me full credits!": Exploring Prompt Injection Attacks on LLM-Based Automatic Grading Systems

arXiv:2606.03090v2 Announce Type: replace-cross Abstract: The emergence of large language models (LLMs) has significantly accelerated recent research on LLM-based automatic grading (AG) systems. Benefiting from the strong instruction-following capabilities and broad prior knowledge of LLMs, educators can deploy AG systems across diverse tasks using only natural language rubrics while achieving satisfactory grading performance. Despite these advantages, new security concerns may also arise. In particular, prompt injection (PI) attacks have recently become a major threat to LLM-based applications. In the context of AG, attackers can potentially exploit PI vulnerabilities to manipulate grading systems into assigning artificially high scores regardless of the actual answer quality. Such behavior poses serious risks to the fairness, reliability, and integrity of educational assessment. In this work, we study PI attacks in AG systems, and systematically investigate the effectiveness of such attacks in educational scenarios. We further evaluate the effectiveness of existing defensive strategies against these attacks. Through comprehensive experiments under rubric-based grading settings, we demonstrate that current LLM-based AG systems remain highly vulnerable to PI attacks. We hope that our findings raise awareness of this emerging threat and motivate future research toward secure, robust, and trustworthy LLM-based educational systems.

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

A Guide to Estimating Conditional Average Treatment Effects in Competing Risks Settings

arXiv:2606.18281v1 Announce Type: cross Abstract: Conditional average treatment effects (CATEs) are central to treatment decision-making in personalized medicine. In competing risks settings, estimating CATEs from survival data allows for patient-specific assessments of treatment effectiveness for a specific event of interest while properly accounting for alternative event types. This distinction is essential in the presence of comorbidities, where competing causes of death may otherwise confound the therapeutic benefit. Focusing on right-censored survival times with binary treatment, we examine CATEs defined as covariate-conditional differences in the absolute risk for the event of interest at a fixed time. To this end, we study meta-learners which adapt machine learning algorithms for CATE estimation in competing risks scenarios. We systematically compare six meta-learners, combining Cox regression or random survival forests for risk modeling with elastic net regression or random forests for direct CATE modeling. To provide practical guidance on model selection, we evaluate their performance in multiple simulation settings, that differ in hazard complexity, treatment heterogeneity, treatment assignment, event type distribution and censoring. To facilitate applied use, we provide the R package, crsurvlearners, which implements all considered approaches.

25.
medRxiv (Medicine) 2026-06-22

GCH1 p.Ser80Asn Confers Risk for Parkinson's Disease in East Asian Populations

Introduction: GCH1 has been implicated in Parkinson's disease (PD), but its risks variants and associations are not well defined. Objectives: To investigate the clinical relevance and PD risk associated with the GCH1 p.Ser80Asn variant. Methods: We first identified a segregating GCH1 p.Ser80Asn variant in a Malaysian Chinese PD family via whole genome sequencing (WGS). We assessed its risk association using multi-ancestry WGS data from the Global Parkinson's Genetics Program (GP2) (n=22,372PD vs n=8,826Controls) and meta-analysis of East Asian (EAS) cohorts (n=4,712PD vs 38,733Controls). Clinico-demographic details of affected variant carriers were collated. Results: The GCH1 p.Ser80Asn variant was enriched in GP2 EAS PD populations (n=9/2,757; 0.33%) but not detected in other ancestries. Meta-analysis revealed increased PD risk in EAS populations (odds ratio:5.1; 95%CI:2.3-10.7; p=2.89x10-5). Affected carriers (mean age at onset:56.3+-12.5 years) had additional occurrence of dystonia, while dementia was rare. Conclusions: The GCH1 p.Ser80Asn variant is a rare, EAS-enriched risk variant for PD.