探索全球前沿学术脉络

01.

medRxiv (Medicine) 2026-06-10 DOI: HASH:816732fd23ded774c994ddac1db3024b

Cortical activity during narrative discourse production in individuals with post-stroke aphasia and controls measured via functional near-infrared spectroscopy

作者:

Braun↗E. J↗Carpenter↗E. A↗Gao↗Yucel↗M. A↗Boas↗D. A↗Kiran↗

Introduction: Aphasia is an acquired language disorder with a significant negative functional impact. Much of the research on aphasia has focused on word-level language comprehension and production. Further evaluation of discourse-level tasks, both at behavioral and neural levels, will allow for an ecologically valid understanding of the functional implications of language impairment in this population. Method: This study evaluated bilateral frontal, temporal, and parietal cortical activity during computer-based narrative production in 14 young neurotypical individuals, 17 individuals with post-stroke aphasia, and 15 age-matched neurotypical participants using functional near-infrared spectroscopy (fNIRS). Oxygenated hemoglobin (HbO) was measured during narrative production following short video clips and compared to HbO during counting aloud. In addition, behavioral measures quantifying in-task performance were correlated with averaged HbO values. Results: Young neurotypical individuals showed greater cortical activity in bilateral language regions for narrative production compared to counting aloud. In contrast, people with aphasia showed positive condition-related effects in the right frontal ROI and the age-matched group showed positive condition-related effects in the left frontal and right precentral ROIs. Each group showed different patterns in relationships between cortical activity and discourse performance measures. Conclusion: Overall, young participants showing more consistent condition-related effects for narrative discourse production than individuals with aphasia and age-matched controls. This study shows the potential for fNIRS to evaluate cortical activity for ecologically valid language tasks in individuals with post-stroke aphasia.

阅读与讨论 → 访问原文 →

02.

bioRxiv (Bioinfo) 2026-06-16 DOI: HASH:41bef19b34f7f7aeeef3f4f18340561f

DMcloud: Macromolecular Structure Modeling Using Local Structure Fitting for Medium to Low Resolution cryo-EM maps

作者:

Terashi↗Wang↗Zhang↗Zhu↗Hong Park↗Kihara↗

Cryogenic electron microscopy (cryo-EM) has become an essential experimental approach in structural biology for determining macromolecular structures. When the resolution of a cryo-EM map is worse than approximately 5[A], fitting known or predicted molecular models into the map becomes a common strategy for interpretation. However, accurately fitting biomolecular models into cryo-EM maps, particularly for large macromolecular complexes, remains challenging when the input structure models contain errors or are in a conformation different from that represented in the map. Here, we present DMcloud, a method for local structure fitting of proteins and nucleic acids in cryo-EM maps. Instead of forcing an entire input model into the map, DMcloud divides input structures into local regions, identifies regions that are supported by the density, removes unsupported regions, and assembles the retained regions into a final model. We benchmarked DMcloud on 176 cryo-EM maps, including intermediate and high-resolution maps that include proteins, DNAs, or RNAs. For EM maps in the 5.0-10.0 [A] and 2.5-5.0 [A] resolution ranges, DMcloud achieved average sequence modeling coverage of 0.49 and 0.70, respectively. For DNA/RNA maps, DMcloud achieved an average sequence coverage of 0.75. Across all datasets, DMcloud consistently outperformed existing methods in model accuracy, map-model correlation, and modeling coverage.

阅读与讨论 → 访问原文 →

03.

arXiv (CS.AI) 2026-06-12 DOI: arXiv:2606.13233

ReSET: Accurate Latency-Critical NVFP4 Reasoning via Step-Aware Temperature Scaling

作者:

Sihwa Lee↗Janghwan Lee↗Donghoon Yoo↗Jae Gon Kim↗Hanyul Ryu↗Soojung Ryu↗Jungwook Choi↗

arXiv:2606.13233v1 Announce Type: cross Abstract: Large reasoning models (LRMs) improve complex problem-solving by generating long intermediate reasoning traces, but this substantially increases inference costs. NVFP4 inference offers a promising approach to reduce both computational and memory costs through hardware-supported low-precision execution. However, directly applying NVFP4 to LRMs introduces two practical limitations: reasoning accuracy degrades under quantization, and existing NVFP4 kernels do not fully realize latency benefits in small-batch autoregressive decoding. In this work, we analyze the effect of NVFP4 quantization on token-level uncertainty during reasoning. We show that quantization increases incorrect sampling at low-entropy symbolic tokens, while causing over-concentration on a small set of tokens in high-uncertainty reasoning steps. Based on this observation, we propose ReSET, a reasoning-step entropy-based temperature-scaling method that estimates step-level uncertainty online and adapts the decoding temperature using both token-level and step-level entropy signals. To address the latency gap, we further design a CUDA-core small-$M$ NVFP4 kernel for latency-critical autoregressive decoding. Across reasoning benchmarks and model scales, ReSET improves NVFP4 reasoning accuracy by up to $\sim\!$2 points over the NVFP4 baseline. Our CUDA-core small-$M$ kernel further improves latency-critical decoding, delivering up to $2.5\!\times$ kernel-level speedup over NVFP4 vLLM and approximately $2\!\times$ end-to-end decoding speedup over BF16. Code is available at https://github.com/aiha-lab/ReSET.

阅读与讨论 → 访问原文 →

04.

arXiv (CS.CL) 2026-06-11 DOI: arXiv:2606.12088

Debiasing Without Protected Attributes: Latent Concept Erasure from Textual Profiles

作者:

Shun Shao↗Zheng Zhao↗Anna Korhonen↗Yftah Ziser↗Shay B. Cohen↗

Most fairness research in NLP assumes direct access to protected attributes such as gender, race, or nationality. In practice, however, such information is often unavailable due to privacy constraints, missing metadata, or legal restrictions, even though models may infer it from indirect textual cues. This raises a key question: can debiasing succeed without direct access to sensitive attributes? We propose H-SAL, which performs post-hoc concept and attribute erasure using self-description text as an implicit debiasing signal. To support this setting, we introduce a multi-domain Stack Exchange-based fairness benchmark for helpfulness prediction that includes both explicit and implicit signals, enabling comparison between standard debiasing with protected labels and debiasing without access to sensitive information. Across encoder and decoder-only language models, we find that implicit self-description often matches or outperforms explicit-label-based debiasing. Our results broaden representation-level fairness research and provide a new benchmark for studying debiasing under realistic data constraints.

阅读与讨论 → 访问原文 →

05.

arXiv (CS.CL) 2026-06-18 DOI: arXiv:2606.18656

The Wrong Kind of Right: Quantifying and Localizing Misfired Alignment in LLMs

作者:

Naihao Deng↗Yiming Feng↗Chimaobi Okite↗Kaijian Zou↗Lu Wang↗Rada Mihalcea↗Yulong Chen↗

Warning: This paper studies stereotypes and biases, and contains potentially disturbing examples, used for illustration purposes only. Our findings should not be interpreted as an argument against alignment. Instead, this paper highlights the need for principled approaches to more advanced alignment. Alignment aims to ensure that large language models (LLMs) behave safely and reliably, including by avoiding unsafe inferences. However, we show that such safety-oriented behaviors can misfire: models may reject warranted conclusions even when they are explicitly supported by context. We call this failure mode misfired alignment, where alignment-induced changes cause LLMs to override explicit evidence. To quantify this phenomenon, specifically on stereotype-related alignment, we introduce VETO, a benchmark consisting of 2,032 BBQ-derived contrastive pairs, and define a new metric, Misfired Alignment Rate (MAR), which measures on a 0 to 100 scale how often a model fails on a stereotype-related question but succeeds on its contrastive counterpart. We benchmark 25 LLMs on VETO, and show that all LLMs, including the most recent ones, exhibit non-trivial (4.7 to 18.9%) MARs while all human participants achieve 0.0% MAR. Controlled priming experiments further show that alignment-induced cues can substantially amplify MAR across LLMs, indicating that these failures are not merely artifacts of individual examples but can be induced by safety-related framing. Mechanistic analyses on open-weight LLMs reveal late-layer suppression of evidence-supported answers, and comparisons between instruct and base LLMs suggest that this suppression emerges after instruction training. These findings show that current alignment methods can overgeneralize surface-level safety cues, to the point of overriding objective evidence, motivating more work on alignment objectives that better preserve contextual grounding.

阅读与讨论 → 访问原文 →

06.

bioRxiv (Bioinfo) 2026-06-22 DOI: HASH:23fb012a78534ae50c8a84cfc3d7038a

CellTosg2Sequence: A Unified Text-Omics-Signaling-Graph Large Language Model for Single-Cell Analysis

作者:

chen↗Ye↗Xu↗Huang↗Zhang↗Li↗Payne↗P. R↗

bioRxivLaTeXUnicodeabstract — In single-cell (sc)-based scientific discovery, text-formatted biomedical prior knowledge and signaling graphs are essential for annotating and interpreting numeric sc-omics data and for generating novel testable hypotheses. A major limitation of existing single-cell large language models (scLLMs) is that they rely on numeric expression data with gene names as the only textual signal, while comprehensive biomedical priors – cellular localization, gene function, disease associations, and signaling interaction patterns – remain absent from the model input. We introduce CellTosg2Sequence, a textual-prior- and signaling-graph-augmented cell-omics-sentence language model. A lightweight heterogeneous graph encoder maps a curated 62,507-node biomedical knowledge graph (KG) into compact virtual tokens that are prepended to each cell sentence, allowing the language model to condition on biological structure with minimal sequence-length overhead. We train CellTosg2Sequence with a three-stage objective: Stage I anchors the KG channel under autoregressive language-model pretraining, leveraging Qwen2.5-32B's own language reasoning for rapid KG alignment; Stage II aligns labels via supervised fine-tuning with KG-anchored InfoNCE; Stage III applies Group Relative Policy Optimization (GRPO) with an ontology-hierarchy reward, enabling free-generation cell-type prediction that generalizes beyond the closed training vocabulary. Across multiple benchmarks and ablation experiments, CellTosg2Sequence outperforms strong baselines. All results are achieved with lightweight LoRA training and a single unified checkpoint.

阅读与讨论 → 访问原文 →

07.

Nature (Science) 2026-06-10 DOI: HASH:e6b46a15e89d32b113385edae094ac79

Measurement of reactor neutrino oscillation with the first JUNO data

作者:

Angel Abusleme↗

Neutrino oscillations (see refs. 1,2 and references therein), a quantum effect manifesting at macroscopic scales, are governed by lepton flavour mixing angles and neutrino mass-squared differences3 that are fundamental parameters of particle physics, representing phenomena beyond the Standard Model. Precision measurements of these parameters are essential for testing the completeness of the three-flavour framework, determining the mass ordering of neutrinos and probing possible new physics. The Jiangmen Underground Neutrino Observatory (JUNO)4 is a 20-ktonne liquid-scintillator detector located 52.5 km from multiple reactor cores, designed to resolve the interference pattern of reactor neutrinos with sub-percent precision5,6. Here we report, using the first 59.1 days of data collected since detector completion in August 2025, the first simultaneous high-precision determination of two neutrino oscillation parameters, $${\sin }^{2}{\theta }_{12}=0.3092\,\pm \,0.0087$$ and $$\Delta {m}_{21}^{2}=(7.50\,\pm \,0.12)\times 1{0}^{-5}\,{\mathrm{eV}}^{2}$$ for the normal mass ordering scenario, improving the precision by a factor of 1.6 relative to the combination of all previous measurements. These results advance the basic understanding of neutrinos, validate the design of the detector and indicate the readiness of JUNO for resolving the neutrino mass ordering with a larger dataset. The rapid achievement with a short exposure highlights the potential of JUNO to push the frontiers of precision neutrino physics and paves the way for its broad scientific programme. The first data of the Jiangmen Underground Neutrino Observatory deliver high-precision neutrino oscillation parameters, improving measurements and demonstrating readiness to determine neutrino mass ordering.

阅读与讨论 → 访问原文 →

08.

arXiv (CS.LG) 2026-06-12 DOI: arXiv:2606.13092

Scale Buys Interpolation, Structure Buys a Horizon: Certified Predictability for Equivariant World Models

作者:

Hongbo Wang↗

arXiv:2606.13092v1 Announce Type: new Abstract: Scale buys interpolation; structure buys a certified horizon. A world model's average error says nothing about whether a particular prediction can be trusted, or for how long. For equivariant latent world models we give a computable, multi-step certificate of the predictable horizon: $T$-step rollout error is provably constant over each symmetry orbit (Theorem A) and stratified channel-by-channel by the predictor's Lyapunov spectrum, $T_j(\epsilon)\sim\log(1/\epsilon)/\lambda_j$. The horizon is two-sided – a matching lower bound makes approximate equivariance provably horizon-limited – and the certificate is exclusive to structure: orbit-constant error characterizes equivariance, so no non-equivariant model has it at any scale. Empirically, on 40-D Lorenz-96 only a $\mathbb{Z}_N$-equivariant network recovers the full Lyapunov spectrum ($R^2{=}0.98$); dense and recurrent baselines fail. Because the spectrum is faithful, the certificate acts, a priori: under a fixed sensing budget a $c\times$-inflated certificate provably needs $c\times$ the budget, and the equivariant certificate meets a budget its inflated dense counterpart cannot – with zero calibration data. The same read-out, unchanged, audits public pretrained world models training-free: TD-MPC2 checkpoints land on the certificate's own scope taxonomy – calibrated where strongly expansive (ratio 0.94-1.02), optimistic where weakly expansive, correctly abstaining where contracting – a map a deployed monitor replicates cell-by-cell, out-of-sample. Across the official 1M-317M multitask ladder, calibration does not improve with parameters. On V-JEPA 2-AC (1B, real robot data) the measured cross-check correctly overrides an over-promising tangent spectrum – the cross-validated audit, not the raw number, is the deployable object. Scale buys interpolation, not a calibrated horizon.

阅读与讨论 → 访问原文 →

09.

arXiv (CS.AI) 2026-06-17 DOI: arXiv:2606.18096

S4oP: Operator-level Pruning of Structured State Space Models for Resource-Constrained Devices

作者:

Marco Deano↗Filippo Ziche↗Nicola Bombieri↗

arXiv:2606.18096v1 Announce Type: cross Abstract: Structured State Space Models (SSMs), including the S4 and S4D architectures, have recently emerged as powerful alternatives to attention-based models for capturing long-range dependencies in sequential data. Despite their strong empirical performance, deploying these models in time- and resource-constrained settings remains challenging due to their computational and memory demands. In this paper, we propose a novel incremental, operator-level pruning approach for S4- and S4D-based models that significantly reduces inference cost while preserving predictive performance. To the best of our knowledge, this is the first work to systematically investigate structured operator pruning for SSMs. Our method progressively prunes model operators by interleaving structured masking with fine-tuning, while jointly monitoring accuracy and inference latency. We implement this approach within a unified training and evaluation framework that enables systematic exploration of efficiency-accuracy trade-offs. Experiments across multiple benchmark datasets show that pruning up to 70% of the model operators preserves the performance of the original models in most cases, while substantially reducing inference latency. These results demonstrate that structured operator pruning is an effective and previously unexplored strategy for improving the efficiency of SSMs and facilitate their deployment in practical, resource-constrained scenarios.

阅读与讨论 → 访问原文 →

10.

arXiv (CS.LG) 2026-06-12 DOI: arXiv:2606.13568

Adjusted Cup-Product Neural Layer

作者:

Snigdha Chandan Khilar↗

arXiv:2606.13568v1 Announce Type: new Abstract: Many important observables in physics and geometry are cup products of cochains. The adjusted cup product neural layer has been introduced in this paper. It is a neural primitive that hard wires the cup product with an adjustment term from higher gauge theory. This creates a readout that is gauge invariant by design. Their main theoretical result shows that on a closed cycle the output relies entirely on the adjustment coefficient. Setting this coefficient to zero removes the output completely regardless of other parameters. Thus the adjustment is the only source of gauge invariant signal. They prove this observable is a nonzero quadratic form and is exactly invariant under one and two gauge transformations.

阅读与讨论 → 访问原文 →

11.

arXiv (CS.CL) 2026-06-11 DOI: arXiv:2606.11648

Dummy Backdoor as a Defense: Removing Unknown Backdoors via Shared Internal Mechanisms for Generative LLMs

作者:

Kazuki Iwahana↗Masaru Matsubayashi↗Takuma Koyama↗Toshiki Shibahara↗Kenichiro Omintato↗Akira Ito↗

Backdoor attacks pose a serious threat to the safety and reliability of Large Language Models (LLMs), as they cause models to behave normally on clean inputs while producing attacker-specified responses when hidden triggers are present. Removing such unknown backdoors is particularly challenging when the defender does not know the backdoor attack types or the internal mechanisms formed through backdoor training. In this work, we propose a simple but effective backdoor removal method based on shared internal mechanisms across different backdoors. First, we show that different backdoors with the same task (attack objective) induce similar trigger-activated changes in the internal activations. Motivated by this observation, our method intentionally embeds a backdoor with a known trigger (dummy backdoor) and then removes it through further fine-tuning on dummy-triggered inputs paired with clean responses. Since the dummy backdoor and the unknown backdoor can rely on shared internal mechanisms, removing the dummy backdoor also reduces the effect of the unknown backdoor. We evaluate our method on three backdoor attack types across multiple model families. Experimental results show that our method substantially reduces the attack success rate of the unknown backdoor while preserving model utility, outperforming representative existing defense methods in both backdoor removal effectiveness and utility preservation. These findings suggest that a defender-controllable backdoor can serve as a helpful proxy for mitigating unknown backdoors in generative LLMs.

阅读与讨论 → 访问原文 →

12.

arXiv (CS.AI) 2026-06-15 DOI: arXiv:2511.07368

Distributional Biases in Post-Training: A Markovian Analysis of Reasoning Trajectories

作者:

Dake Bu↗Wei Huang↗Andi Han↗Atsushi Nitanda↗Bo Xue↗Qingfu Zhang↗Hau-San Wong↗Taiji Suzuki↗

arXiv:2511.07368v3 Announce Type: replace-cross Abstract: Foundation models exhibit broad knowledge but limited task-specific reasoning, motivating post-training strategies such as RL with verifiable rewards (RLVR) and test-time scaling (TTS). While recent work highlights the role of exploration in improving pass@K, empirical evidence points to a paradox: RLVR and ORM/PRM typically reinforce existing paths rather than expanding the reasoning scope, raising the question of why exploration helps if no new patterns emerge. To reconcile this paradox, we adopt the perspective of Kim et al. (2025), viewing easy (e.g., simplifying a fraction) versus hard (e.g., discovering the some symmetry) reasoning steps as low versus high probability Markov transitions. In this tractable model, pretraining corresponds to tree-graph discovering, while post-training corresponds to CoT reweighting. We provably show that, both RLVR and ORM/PRM would favor heavily to several high-probability paths, and thereby forget rare-but-crucial CoTs. Building on this, we further prove that exploration strategies such as rejecting easy instances and KL regularization help preserve rare CoTs. Empirical simulations corroborate our theoretical results.

阅读与讨论 → 访问原文 →

13.

arXiv (CS.LG) 2026-06-18 DOI: arXiv:2606.18431

Beyond Prediction: Tail-Aware Scheduling for LLM Inference

作者:

Yueying Li↗Yuanfan Chen↗Jiayang Chen↗Esha Choukse↗Haoran Qiu↗G. Edward Suh↗Rodrigo Fonseca↗Ziv Scully↗Udit Gupta↗

arXiv:2606.18431v1 Announce Type: new Abstract: LLM serving exhibits extreme length variability, making size-based scheduling difficult in practice. Recent LLM schedulers approximate SJF/SRPT using predicted decode lengths or ranks and primarily report mean-centric metrics such as TTFT and TBT. We show that these prediction-driven policies can be fragile under distribution shifts, bursty arrivals, and GPU memory pressure, while offering limited control over the tail latency (P90-P99) that dominates user experience, even with perfect decode-length knowledge. We introduce a distribution-aware, prediction-free scheduling framework that replaces explicit length prediction with soft priority boosting driven by lightweight statistical signals. Our design co-optimizes scheduling and cache-aware preemption to account for memory-coupled decode dynamics across workload mixes. Evaluated on production and open-source traces, our method reduces P99 TTLT by up to 35-50% relative to SRPT with perfect length knowledge and reduces TTFT by 34-47% across workloads, including reasoning-heavy and chat-heavy tasks. These results demonstrate a robust alternative for optimizing tail latency in online LLM serving.

阅读与讨论 → 访问原文 →

14.

arXiv (CS.AI) 2026-06-11 DOI: arXiv:2606.10120

MetaPlate: Counterfactual-Guided RAG-LLM Tool for Personalized Food Recommendation and Hyperglycemia Prevention

作者:

Asiful Arefeen↗Carol Johnston↗Hassan Ghasemzadeh↗

arXiv:2606.10120v2 Announce Type: replace-cross Abstract: Postprandial hyperglycemia is a key risk factor for metabolic disorders; however, existing dietary guidance is often static, impractical, and insufficiently personalized, providing recommendations that are difficult to follow or not impactful. While recent advances leverage continuous glucose monitoring (CGM) and machine learning to predict glycemic responses, these approaches are largely predictive and lack actionable guidance. Moreover, recommendation systems are often misaligned with user goals and require extensive input. We present MetaPlate, a counterfactual explanation (CF) guided, context-aware decision-support framework that generates personalized meal recommendations to mitigate postprandial glucose excursions in healthy adults. MetaPlate integrates multimodal data, including CGM readings, wearable-derived physiological signals, and user-provided meal inputs from $25$ individuals to model pre-meal context. A machine learning model predicts glucose response, while a CF optimization module adjusts meal composition modifying macronutrient amounts to maintain glucose levels within a target range ($\leq 140$ mg/dL). An LLM-based retrieval-augmented generation (RAG) layer enhances interpretability by producing human-readable recommendations using constrained search of the USDA food database. We evaluate MetaPlate via a structured expert-in-the-loop assessment with registered dietitians (RDs), comparing performance before and after prompt refinement. Results show improvements in meal realism, portion suitability, and recommendation likelihood, with expert feedback indicating a shift from clinically implausible outputs to actionable, contextually appropriate recommendations. Our findings emphasize the importance of domain knowledge and structured constraints in LLM-driven systems and highlight the potential of MetaPlate as a real-time personalized dietary decision-support tool.

阅读与讨论 → 访问原文 →

15.

arXiv (CS.AI) 2026-06-15 DOI: arXiv:2505.17961

Federated Causal Inference from Multi-Site Observational Data via Propensity Score Aggregation

作者:

R\'emi Khellaf↗Aur\'elien Bellet↗Julie Josse↗

arXiv:2505.17961v4 Announce Type: replace-cross Abstract: Causal inference typically assumes centralized access to individual-level data. Yet, in practice, data are often decentralized across multiple sites, making centralization infeasible due to privacy, logistical, or legal constraints. We address this problem by estimating the Average Treatment Effect (ATE) from decentralized observational data via a Federated Learning (FL) approach, allowing inference through the exchange of aggregate statistics rather than individual-level data. We propose a novel method to estimate propensity scores via a federated weighted average of local scores using Membership Weights (MW), defined as probabilities of site membership conditional on covariates. MW can be flexibly estimated with parametric or non-parametric classification models using standard FL algorithms. The resulting propensity scores are used to construct Federated Inverse Propensity Weighting (Fed-IPW) and Augmented IPW (Fed-AIPW) estimators. In contrast to meta-analysis methods, which fail when any site violates positivity, our approach exploits heterogeneity in treatment assignment across sites to improve overlap. We show that Fed-IPW and Fed-AIPW perform well under site-level heterogeneity in sample sizes, treatment mechanisms, and covariate distributions. Theoretical analysis and experiments on simulated and real-world data demonstrate clear advantages over meta-analysis and related approaches.

阅读与讨论 → 访问原文 →

16.

arXiv (CS.CV) 2026-06-16 DOI: arXiv:2606.14777

JoyAI-VL-Interaction: Real-Time Vision-Language Interaction Intelligence

作者:

Dingyu Yao↗Junhao Zhou↗Chenxu Yang↗Chuanyu Qin↗Haowen Hou↗Zheming Liang↗Congcong Wang↗Yuhang Cao↗Shenglong Ye↗Shuai Xie↗Shuhuan Gu↗Haoyang Huang↗…

Many moments in the real world do not wait for a user to ask. A fire starts on a security monitor, an expression flickers across a video call, or a product a viewer wants flashes by in a livestream. Yet today's large models remain mostly turn-based by design: they answer only when addressed, and even video-call apps that appear interactive still operate as question-answer systems, reacting only when polled or prompted. We argue for a different paradigm: a model that is present in the world like a person. It continuously watches what is happening now, decides on its own whether to speak or stay silent, interacts in real time, and delegates to a background model when the problem is hard. To advance interaction models and their adoption across domains, we make two fully open-sourced contributions. First, we release JoyAI-VL-Interaction, an 8B-scale, vision-first VL-interaction model. The model makes the response decision internally, choosing each second to stay silent, respond, or delegate to a background model, and it excels at vision-triggered responsiveness and time awareness. We pair it with a transferable training recipe, from which capabilities we never trained for emerge, such as guiding a shopper through changing app screens or improvising a lecture from a slide deck. Second, we release a complete, deployable system built around that model. The system streams any ongoing video into the model, making it genuinely present in the world. All other components are pluggable, including ASR/TTS modules, memory, visualization UI, and a background brain that can connect to any API or agent. Across six real-world scenarios, human raters prefer JoyAI-VL-Interaction over the in-app video-call assistants of Doubao and Gemini by a wide margin. To our knowledge, this is the first open, vision-driven interaction model released together with its training recipe, data, and complete deployable system.

阅读与讨论 → 访问原文 →

17.

arXiv (CS.CV) 2026-06-12 DOI: arXiv:2606.12655

Amnesia: A Stealthy Replay Attack on Continual Learning Dreams

作者:

Ahmed Sharshar↗Naveen Kumar Kummari↗Mohsen Guizani↗

Continual learning (CL) models often use experience replay to reduce catastrophic forgetting, but their robustness to replay sampling interference remains underexplored. Existing CL attacks alter inputs or training pipelines (poisoning/backdoors) and rarely include explicit auditable constraints, limiting realism. Here, auditability means a monitor can verify compliance from sampler-visible telemetry - e.g., logged replay index/label statistics - by checking that the realized replay class histogram stays close to a nominal baseline and that replay rate is unchanged per batch and/or over a rolling window. We study a limited-privilege insider who controls only replay index selection, not pixels, labels, or model parameters, while staying within auditable limits such as queue priorities. We introduce Amnesia, a replay composition attack that maximizes degradation under two budgets: a visibility budget delta bounding the TV/KL divergence from a nominal class histogram p0, and a mass budget f fixing the replay rate. Amnesia has two steps: (i) compute lightweight class utilities, such as EMA loss or confidence, to tilt p0 toward harmful classes; and (ii) project the tilt back into the delta-ball using efficient KL (exponential tilt) or TV (balanced mass redistribution) optimizers. A windowed scheduler enforces rolling audits. Across challenging CL benchmarks and strong replay baselines, Amnesia consistently lowers final accuracy (ACC) and worsens backward transfer (-BWT). The KL variant delivers high impact while remaining largely undetected under multiple audit schemes, including per-batch and rolling-window checks. The TV variant is more damaging but easier to detect, especially under tight per-class constraints. These results expose index-only replay control as a practical, auditable threat surface in CL systems and establish a principled impact-visibility trade-off.

阅读与讨论 → 访问原文 →

18.

arXiv (CS.LG) 2026-06-19 DOI: arXiv:2605.09609

Minimal Filling Architectures of Polynomial Neural Networks: Counterexamples, Frontier Search, and Defects

作者:

Kevin Dao↗Jose Israel Rodriguez↗

arXiv:2605.09609v2 Announce Type: replace Abstract: We provide counterexamples to the unimodal minimal filling architecture conjecture for polynomial neural networks (PNNs) with power activation functions. Fixing the input and output widths, the conjecture states that any minimal filling architecture has unimodal widths for the hidden layers. We found counterexamples via a frontier search, recursive dimension bounds on neurovarieties, and symbolic computation. Notably, several subarchitectures of our main example exhibit large defect, in contrast with the predominantly small-defect behavior observed in prior literature.

阅读与讨论 → 访问原文 →

19.

arXiv (CS.AI) 2026-06-16 DOI: arXiv:2606.14923

Trust Between AI Agents: Measuring Formation, Breakage, and Recovery, with Implications for Governing Multi-Agent Systems

作者:

Yujiao Chen↗

arXiv:2606.14923v1 Announce Type: new Abstract: As language-model agents increasingly work in teams, each agent must decide how much to trust its teammates. Yet we lack a standard way to measure trust between AI agents. We propose a behavioral measure based on costly verification. In a cooperative survival game, checking a teammate's work consumes resources, while trusting a wrong answer can be fatal. Relative to a memoryless version of the same model, reduced verification provides an observable measure of trust. Using this framework, we study trust formation, breakage, and recovery across six frontier model snapshots. When paired with a consistently reliable teammate, four snapshots (Claude Opus 4.6, Claude Sonnet 4.6, GPT-5.1, and Gemini 3.1 Pro) reduce verification by roughly 60-85%, whereas two smaller snapshots show little or no such adjustment. Failures reverse this discount, but models differ in how they respond. Some concentrate renewed scrutiny on the culprit, while others become more cautious toward the entire team. Recovery is slower than formation, and clustered failures sustain suspicion far longer than the same number of failures spread apart. These differences have practical consequences. Models that form trust verify less, decide more quickly, and achieve higher payoffs in our environment. By contrast, persistent over-verification is associated with indecision rather than safety. Our results show that trust dispositions can be measured before deployment and suggest that calibration, rather than maximal suspicion, should be the central concern in the governance of multi-agent AI systems.

阅读与讨论 → 访问原文 →

20.

arXiv (CS.AI) 2026-06-11 DOI: arXiv:2606.11215

The Environmental Cost of LLMs in AIED: Reporting and Practices

作者:

Sabrina C. Eimler↗Lukas Erle↗Daniel Flood↗Aditi Haiman↗Luca H\"ackert↗Andr\'e Helgert↗Lachlan McGinness↗B\"usra Yapici↗

arXiv:2606.11215v1 Announce Type: cross Abstract: Large Language Model (LLM) usage in recent years has become increasingly widespread in the Artificial Intelligence in Education (AIED) community. While LLMs offer unique avenues for learners and educators, using LLMs comes with computational and environmental costs. These costs are mostly hidden due to a lack of standardised procedures to measure and report these impacts. To address this gap, we first conducted a literature review of all papers published as part of the AIED 2025 conference proceedings, determining if and how computational or environmental costs of LLMs are reported. Most projects use LLMs, but few report computational resources used and almost none discuss environmental impacts of LLMs as an ethical concern. To address this lack of standardised reporting practices, we propose an open-source method for systematically measuring and reporting the computational expense of LLMs and environmental impact of running Machine Learning (ML) AIED systems. We provide software solutions to measure the carbon footprint for both local and cloud based hardware. We also provide an easy-to-use formula to calculate the computational expense of frontier LLMs even when the exact number of parameters is not known. Overall, we hope to motivate colleagues to use our method to strive for more transparent reporting of hidden costs of using LLMs in the AIED community.

阅读与讨论 → 访问原文 →

21.

arXiv (CS.AI) 2026-06-11 DOI: arXiv:2606.12240

Multi-Rate Mixture of Experts for Accelerating Liquid Neural Network Training

作者:

Shilong Zong↗Almuatazbellah Boker↗Hoda Eldardiry↗

arXiv:2606.12240v1 Announce Type: cross Abstract: Multivariate time-series data often exhibit complex temporal dependencies, irregular sampling, and heterogeneous dynamics across multiple time scales, making accurate sequence modeling particularly challenging. Traditional recurrent neural networks (RNNs), such as Long Short-Term Memory (LSTM) networks, operate in discrete time and may struggle to effectively capture continuous and irregular temporal behaviors. Liquid Neural Networks (LNNs) address some of these limitations through continuous-time dynamics, but standard LNN architectures typically rely on a single dynamical system, limiting their ability to model heterogeneous temporal patterns. To address these challenges, we propose a Multi-Rate Mixture-of-Experts (MR-MoE) framework built on top of Liquid Neural Networks. In the proposed architecture, multiple LNN-based experts operate at distinct time scales, enabling the model to explicitly separate fast-changing dynamics from slow-evolving temporal trends. A gating network further enables adaptive expert specialization based on input conditions. In addition, we incorporate both feature-level and temporal attention mechanisms to improve robustness, interpretability, and long-range dependency modeling. Feature-level attention suppresses noisy or irrelevant variables, while temporal attention selectively focuses on informative historical states. We evaluate the proposed framework on a complex multivariate time-series prediction task and compare it against strong baselines, including LSTM, monolithic LNN, and standard MoE models. Experimental results demonstrate that the proposed MR-MoE framework consistently achieves improved AUROC and AUPRC performance while maintaining favorable computational efficiency. These results highlight the effectiveness of combining continuous-time dynamics, multi-scale expert decomposition, and adaptive attention mechanisms for time-series modeling.

阅读与讨论 → 访问原文 →

22.

arXiv (quant-ph) 2026-06-16 DOI: arXiv:2505.23893

A complexity theory for non-local quantum computation

作者:

Andreas Bluhm↗Simon H\"ofer↗Alex May↗Mikka Stasiuk↗Philip Verduyn Lunel↗Henry Yuen↗

arXiv:2505.23893v2 Announce Type: replace Abstract: Non-local quantum computation (NLQC) replaces a local interaction between two systems with a single round of communication and shared entanglement. Despite many partial results, it is known that a characterization of entanglement cost in at least certain NLQC tasks would imply significant breakthroughs in complexity theory. Here, we avoid these obstructions and take an indirect approach to understanding resource requirements in NLQC, which mimics the approach used by complexity theorists: we study the relative hardness of different NLQC tasks by identifying resource efficient reductions between them. Most significantly, we prove that $f$-measure and $f$-route, the two best studied NLQC tasks, are in fact equivalent under $O(1)$ overhead reductions. This result simplifies many existing proofs in the literature and extends several new properties to $f$-measure. For instance, we obtain sub-exponential upper bounds on $f$-measure for all functions, and efficient protocols for functions in the complexity class $\mathsf{Mod}_k\mathsf{L}$. Beyond this, we study a number of other examples of NLQC tasks and their relationships.

阅读与讨论 → 访问原文 →

23.

arXiv (CS.CL) 2026-06-15 DOI: arXiv:2606.13715

WorkBench Revisited: Workplace Agents Two Years On

作者:

Olly Styles↗

The best agent on WorkBench in March 2024, GPT-4, completed 43% of tasks and took an unintended harmful action, such as emailing the wrong person, on 26% of them. We re-visit the benchmark in June 2026 and find that the best agent to date, Claude Opus 4.8, completes 89% and takes an unintended harmful action on 2.5%. Aside from this considerable progress in frontier agent performance, three things stand out. First, capability and safety go together on WorkBench rather than trade off, so the models that finish the most tasks also do the least unintended damage. Second, while several classes of error have been totally eliminated, frontier models still make some basic mistakes that occasionally result in irreversible harm, such as sending an email to the wrong person. Third, the rise of open-weight models has drastically lowered costs for a performance level that was previously only accessible to proprietary models, while frontier costs have stayed relatively stable. We release an updated version of the benchmark with data and code quality improvements, new model scores, and analysis of agent progress on WorkBench since 2024.

阅读与讨论 → 访问原文 →

24.

arXiv (CS.AI) 2026-06-17 DOI: arXiv:2606.17930

How Inference Compute Shapes Frontier LLM Evaluation

作者:

Jessica McFadyen↗Ole Jorgensen↗Harry Coppock↗Kevin Wei↗Cozmin Ududec↗

arXiv:2606.17930v1 Announce Type: new Abstract: AI evaluations are shifting toward harder tasks that benefit from longer trajectories involving tool use and iterative problem solving. As a result, performance is increasingly sensitive to the amount and allocation of compute available at test time ("inference compute"). Yet many evaluations still report performance at a single restrictive budget, meaning that low scores may reflect the evaluation setup rather than the model's underlying capability. To test this, we evaluate up to 12 frontier language models on seven challenging benchmarks spanning software engineering, mathematics, medicine, and cybersecurity. We use a controlled setup combining three simple inference-scaling interventions: larger token budgets, context compaction, and repeated submission attempts, guided either by the model itself or by minimal correctness feedback. We find three main results. First, larger token budgets substantially improve performance on benchmarks across multiple domains, including cybersecurity, FrontierMath, Humanity's Last Exam, and TerminalBench. Second, fixed-budget evaluations can increasingly understate frontier capability as models advance. Newer models reach higher performance at large budgets, where they unlock harder tasks and solve them more reliably. Third, benchmarks differ in which inference-scaling methods help most: repeated submission broadly improves performance, but the value of larger token budgets, external feedback, and parallel attempts varies by benchmark. Overall, our results show that benchmark scores are protocol-dependent. We therefore argue that evaluations should report capability as a function of inference-time compute, specify protocol choices explicitly, and compare model generations over a large shared compute range at matched budgets, especially in safety- or policy-relevant settings.

阅读与讨论 → 访问原文 →

25.

arXiv (CS.LG) 2026-06-15 DOI: arXiv:2606.14215

LapidaryEngine: Fully Conversational Crystal Generation

作者:

Yusei Ito↗Yuta Suzuki↗Tomoya Murata↗Masaki Adachi↗

arXiv:2606.14215v1 Announce Type: new Abstract: The emergence of Large Language Models (LLMs) has inspired the vision of generating bespoke crystal materials directly from natural-language instructions, enabling users to design materials through intuitive, conversational interaction. Existing text-to-crystal generative models represent important early steps toward this goal, but they suffer from two critical limitations: (i) restricted input formats that require highly structured descriptions (e.g., chemical formulas), and (ii) one-directional generation, where models can map text to crystal but cannot perform the inverse. These limitations prevent fully conversational workflows and hinder alignment with users' inherently ambiguous and evolving desiderata. We address these challenges with LapidaryEngine, the first model to support fully conversational crystal generation. LapidaryEngine accepts free-form natural-language requests and performs iterative refinement and editing in a dialogue-like manner. The key innovation is a pivot representation, a third, intermediate form that enables bidirectional translation between text and crystal structures despite the absence of direct paired datasets. Leveraging this pivot allows robust interpretation of user feedback and precise structural control. We demonstrate LapidaryEngine across diverse tasks, including insulator discovery, stability optimization, compositional modification, and structural editing, showcasing its ability to align generated materials with user intent in an interactive manner.

阅读与讨论 → 访问原文 →