论文广场 - AcademicHub

01.

arXiv (quant-ph) 2026-06-15 DOI: arXiv:2606.14447

Dealing with locality in QAOA

作者:

arXiv:2606.14447v1 Announce Type: new Abstract: Shallow-depth QAOA on sparse, high-diameter MaxCut instances faces a locality bottleneck: at depth $p$, local observables can depend only on a bounded neighborhood of the circuit interaction graph. We propose a transport-augmented QAOA that keeps the MaxCut cost Hamiltonian unchanged but enriches the mixer with optimized, unweighted shortcut couplings (scheduled $XX+YY$) to collapse the effective interaction-graph diameter. Using exact finite-depth support recursions, we relate optimal shortcut placement to bounded-diameter graph augmentation, and show in benchmarks that (unlike ma-QAOA) performance becomes effectively size-invariant once the diameter is reduced. For bipartite families (base diameter 4), reducing the interaction path to $d=1$ raises the ensemble-averaged approximation ratio from 0.7378 (ma-QAOA) to 0.9767 at $p=1$ ($\sigma=0.0251$, nine system sizes); on random trees (base diameter 10), at $p=2$ it improves from 0.9226 to 0.9997 ($\sigma=0.0001$).

阅读与讨论 → 访问原文 →

02.

medRxiv (Medicine) 2026-06-15 DOI: HASH:d35f5546b1136146ea9d698406195a78

High Demand, Low Possession: Dilemmas and Strategies for Research Capability Cultivation in Clinical Medicine Postgraduates

作者:

Wang ↗Yang ↗Gong ↗

Most previous studies have examined medical postgraduate research training from a single dimension, lacking a full-chain analysis that integrates capability demand, actual possession, obstacles, and output. Consequently, the measurement of capability gaps and the analysis of underlying training model deficiencies remain insufficient. To address this gap, we administered a self-designed multidimensional questionnaire to 86 clinical medicine postgraduates at a medical school, covering research cognition, interest, capability demand and possession, participation pathways, difficulties, and outputs. The aim was to systematically characterize the current situation, identify problems, and propose optimization strategies. Over 90% of participants expressed interest in research, yet only 1.16% self-rated as very knowledgeable. The largest demand-possess gap was for writing and publication (86.05% vs. 16.28%), followed by independent research capability (75.58% vs. 11.63%). A total of 59.30% cited lack of foundational knowledge, making experiments very difficult, as the greatest challenge, and 66.28% had no research achievements. The primary source of research topics was supervisor assignment (54.65%), with only 4.65% choosing topics independently. No statistically significant differences were found across grades or training types (P > 0.05). These findings reveal a structural high demand, low possession gap in medical postgraduate research training, with early research experience deficit and a passive research model as key constraining factors. Accordingly, an integrated bachelor-postgraduate progressive research competency training system is proposed.

阅读与讨论 → 访问原文 →

03.

arXiv (CS.CL) 2026-06-16 DOI: arXiv:2606.16576

Can LLM Agents Infer World Models? Evidence from Agentic Automata Learning

作者:

Reef Menaged ↗Gili Lior ↗Shauli Ravfogel ↗Roee Aharoni ↗Gabriel Stanovsky ↗

We propose agentic automata learning to evaluate the extent to which tool-calling LLM agents can uncover hidden environments through interaction. In our setup, an agent should uncover a hidden deterministic finite automaton (DFA) by interacting with an oracle through (1) membership queries ("Does this string belong to the target language?") and (2) equivalence queries ("Is this the target DFA?"). This yields a scalable testbed with controlled task complexity, measurable interaction efficiency, and strong baselines (classic automata-learning algorithms). Evaluating state-of-the-art LLMs, we find that performance drops sharply as DFA size increases. Reasoning models are markedly stronger than non-reasoning models, yet trajectory analyses reveal recurring failures in query planning, evidence integration, and hypothesis construction. Overall, our results show that current LLM agents can sometimes perform non-trivial interactive discovery, but remain far less robust and efficient than classic algorithms for the task.

阅读与讨论 → 访问原文 →

04.

arXiv (CS.AI) 2026-06-16 DOI: arXiv:2606.15291

A Formal Framework for Declarative Agentic AI in Business Process Analysis

作者:

Mohammad Azarijafari ↗Luisa Mich ↗Michele Missikoff ↗

arXiv:2606.15291v1 Announce Type: new Abstract: Agentic AI opens new opportunities for automating Business Process (BP), enabling autonomous decision-making and dynamic adaptation. However, realising this potential requires BP entities and their interactions to be defined with formal precision. This paper presents a formal framework for Agentic BP analysis through the AGO methodology. AGO captures the modelling perspective in terms of who is acting (Agents), why it is carried out (Goals), and what the relevant entities are (Objects). Grounded in set theory and mathematical logic, we formally define the AGO entity types and their interactions, organising all definitions into a BP Knowledge Base (BPKB). The resulting BPKB supports structured querying, incremental updates, and automatic generation of BP workflows, while ensuring soundness and completeness of the derived paths.

阅读与讨论 → 访问原文 →

05.

Nature (Science) 2026-06-17 DOI: HASH:2a492025d8da341386b002dd9a4f8a5d

How the brain builds sentences, neuron by neuron

作者:

Max Kozlov ↗

Neural maps reveal the specialized cells that produce speech. Neural maps reveal the specialized cells that produce speech.

阅读与讨论 → 访问原文 →

06.

arXiv (CS.AI) 2026-06-17 DOI: arXiv:2606.17735

Shattering the Autoregressive Curse: Dynamic Epistemic Entropy Orchestrated Erasable Reinforcement Learning for LLMs

作者:

Ziliang Wang ↗Kang An ↗Faqiang Qian ↗Jialu Cai ↗Cijun Ouyang ↗Yuhang Wang ↗Qibing Ren ↗Yichao Wu ↗

arXiv:2606.17735v1 Announce Type: new Abstract: Although reinforcement learning (RL) has expanded the cognitive boundaries of large language models (LLMs), it often remains vulnerable to the autoregressive curse in long-horizon logical reasoning: small epistemic perturbations introduced early in generation can propagate irreversibly along the Markov decision process flow, triggering cascading failures that drive the reasoning trajectory toward collapse. To overcome this autoregressive cascade, in which a single early mistake can compromise all subsequent reasoning steps, we propose dynamic epistemic entropy orchestrated erasable reinforcement learning ($E^3RL$). $E^3RL$ eliminates reliance on external signals by grounding the model's endogenous local autoregressive cross-entropy as an intrinsic coordinate of epistemic uncertainty. By introducing segment-level adaptive dynamic thresholds and advantage allocation, $E^3RL$ enables the model to precisely excise localized logical defects while reusing historical key-value (KV) cache streams, thereby endowing the reasoning process with a self-healing capability. We train $E^3RL$ on the DeepMath-103k dataset. Experimental results show that $E^3RL$ reshapes the exploration efficiency of long-sequence reasoning and improves sample efficiency while maintaining linear memory overhead. On mathematical reasoning benchmarks such as AIME, $E^3RL$ achieves substantial performance gains, with the 4B and 8B parameter models surpassing previous state-of-the-art (SOTA) results by 5.349\% and 6.514\%, respectively. These findings suggest that $E^3RL$ shatters the autoregressive curse in long-sequence reasoning and establishes a theoretical and systems-level foundation for the next generation of self-healing artificial general intelligence (AGI).

阅读与讨论 → 访问原文 →

07.

bioRxiv (Bioinfo) 2026-06-18 DOI: HASH:7fcefcd040c1145a35d06e6a4578f1ed

Benchmarking attention-based methods for vision transformers' interpretability in retinal fundus imaging

作者:

Bors ↗Beyeler ↗Trofimova ↗VascX Consortium ↗Presby ↗Bontempi ↗Bergmann ↗

Deep learning models based on Vision Transformers (ViTs) have shown strong performance in retinal fundus imaging, but their interpretability remains poorly understood. In particular, attention-based attribution methods are widely used to explain ViT predictions, despite limited evaluation of their faithfulness and biological relevance in medical imaging. Here, we systematically benchmark four attention-based interpretability methods for RETFound, a retinal ViT-based foundation model, that we previously fine-tuned to predict 17 retinal vascular phenotypes from UK Biobank fundus images1. We compare raw attention, attention rollout, gradient-weighted attention rollout, and Chefer's hybrid relevance-based method using both qualitative visualisation and quantitative evaluation frameworks. To assess attribution faithfulness, we perform perturbation-based deletion and insertion experiments, quantifying changes in model predictions as highly attended image regions are progressively removed or restored. To evaluate biological specificity, we run structure-aware analyses combining attribution maps with vessel segmentation and artery-vein labels through the Relative ratio of Attention Intensity (RAI) metric. Across models, attribution maps differed substantially depending on the selected interpretability method, highlighting the need for rigorous quantitative evaluation. Among the evaluated approaches, gradient-weighted attention rollout consistently achieved the strongest perturbation performance and produced attribution maps most closely aligned with the anatomical definition of the predicted retinal traits. Furthermore, vessel-type specific models systematically concentrate attention on the corresponding vascular structures despite being trained using only a single scalar value per image as supervision. These findings demonstrate that attention-based attribution methods capture biologically meaningful vascular representations, while also revealing method-dependent variability in attribution behaviour. This work provides a quantitative framework for evaluating interpretability methods in medical imaging with annotated segmentation and contributes toward more transparent and biologically grounded medical AI systems.

阅读与讨论 → 访问原文 →

08.

arXiv (CS.CV) 2026-06-15 DOI: arXiv:2606.13911

Overhead Wildlife Locator (OWL): Benchmarking Weakly Supervised Learning for Aerial Wildlife Surveys

作者:

Automated aerial wildlife surveys increasingly rely on deep learning, yet standard object detectors require bounding-box annotations, reported to be up to seven times slower and three times more expensive to produce than point-level labels. To address this bottleneck, we introduce the Overhead Wildlife Locator (OWL), a weakly supervised density-estimation framework with three variants: OWL-C, a fully convolutional model for high-throughput screening; OWL-T, a Swin-augmented hybrid for heterogeneous, cluttered scenes; and OWL-D, built on a frozen DINOv3 ViT-H+/16 encoder with a DPT-style fusion decoder. We benchmark all three against POLO, YOLOv11n, and YOLOv11l across five public aerial datasets, from sparse fixed-wing savanna surveys to dense UAV paddock imagery, and against the published HerdNet baseline on its native Delplanque split. OWL-D sets a new state of the art on Delplanque (0.934 AP vs. HerdNet's 0.840) and records the highest AP on four of the five datasets. Performance is regime-dependent: on the extreme-density SheepCounter UAV dataset the hybrid OWL-T leads (0.978 AP) and the convolutional variants attain the lowest counting error, whereas the foundation-based OWL-D degrades, indicating which variant suits which survey type. We further validate operational readiness on the Alaska Department of Fish and Game's 2022 Central Arctic Caribou census: under cross-herd and cross-temporal transfer, OWL-C fine-tuned on the 2017 Porcupine Caribou Herd split attains F1 = 0.965 on a held-out patch test set, with a signed count error of +3.1% aggregated across the released test patches. We release the OWL code, model weights, and the annotated Porcupine Caribou Herd 2017 (PCH) and Central Arctic Herd 2022 (CAH) patches, the first open patch-level datasets for large-scale caribou aerial surveys, at https://github.com/microsoft/MegaDetector-Overhead.

阅读与讨论 → 访问原文 →

09.

arXiv (CS.AI) 2026-06-18 DOI: arXiv:2606.19079

ARIADNE: Agnostic Routing for Inference-time Adapter DyNamic sElection

作者:

Enrico Cassano ↗Micha{\l} Brzozowski ↗Zuzanna Dubanowska ↗Paolo Mandica ↗Neo Christopher Chung ↗

arXiv:2606.19079v1 Announce Type: new Abstract: The increasing deployment of parameter-efficient fine-tuning (PEFT) has led to model ecosystems in which a single backbone is paired with many task-specialized adapters. In this setting, inference-time queries often arrive without task labels, requiring the system to automatically select the most appropriate adapter from a growing and heterogeneous adapter pool. Existing routing methods either depend on access to adapter internals, such as weight decompositions or gradient-based statistics, or require additional router training, which limits scalability and portability as new adapters are added. We introduce ARIADNE, a training-free, adapter-agnostic routing framework for dynamic adapter selection at inference time. ARIADNE represents each adapter through a set of centroids computed from embeddings of its training set, capturing the data distribution associated with that adapter. Given an unlabeled input, it selects an adapter by measuring proximity to these centroids in latent space. Because routing is performed entirely in the input embedding space, ARIADNE is compatible with arbitrary PEFT methods and requires no modification to the adapters or training procedures. Primarily evaluated with Llama 3.2 1B Instruct on 23 diverse NLP tasks, ARIADNE recovers 97.44% of the upper bound performance. Scaling to 44 tasks, it achieves 89.7% average selection accuracy, without additional training or access to adapter internals.

阅读与讨论 → 访问原文 →

10.

arXiv (math.PR) 2026-06-16 DOI: arXiv:2606.16636

Free energy of non-convex multi-species spin glasses with centered Ising spins

作者:

Hong-Bin Chen ↗Victor Issa ↗Jean-Christophe Mourrat ↗

arXiv:2606.16636v1 Announce Type: new Abstract: We identify the limit free energy of all multi-species spin glasses with centered $\pm 1$ spins. The result was previously known only under a convexity assumption on the covariance function of the Hamiltonian. We also obtain a one-species reduction of the formula for balanced multi-species models.

阅读与讨论 → 访问原文 →

11.

arXiv (CS.CL) 2026-06-11 DOI: arXiv:2606.12117

Soft-Prompt Tuning for Fair and Efficient LLM Benchmark Evaluation

作者:

Selen Erkan ↗Bastian Boll ↗Kristian Kersting ↗Bj\"orn Deiseroth ↗Letitia Parcalabescu ↗

Benchmark scores often misrepresent a large language model's (LLM's) knowledge, because they rely, e.g., on the model's ability to follow specific formatting requirements. This especially penalizes base models that may know the correct answers but lack the ability – typically introduced in post-training – to structure them as instructed. To overcome this, we propose soft-prompt tuning, an efficient, fair, and architecture-agnostic model evaluation. By optimizing only 10 soft-prompt vectors (roughly 0.0006% parameters for a 7B model) over a short tuning period, we adapt models to specific benchmark formats, closing gaps in format-following and ensuring that underlying knowledge is accurately reflected in benchmark scores. This allows one to fairly compare different base models – trained with various pre-training recipes – on benchmarks without the need for full post-training. We evaluated soft-prompt tuning across 7 models and 7 datasets. The results show that (a) soft-prompt tuning saturates format-following within 80 steps (~640 samples) making it highly efficient, (b) soft-prompt tuning significantly outperforms zero- and few-shot prompting, surfacing base model knowledge that standard prompting misses, that (c) even post-trained models can benefit from soft-prompts to maximize format compliance, and that (d) soft-prompted base model performance predicts post-trained model rankings more reliably than zero- and few-shot baselines, offering a low-cost proxy for downstream model quality. Our contributions include (1) metrics which disentangle format-following and knowledge accuracy, (2) a fairer benchmarking protocol of LLM knowledge, and (3) a cost- and memory-effective recipe to identify optimal pre-training strategies early in LLM development.

阅读与讨论 → 访问原文 →

12.

arXiv (CS.CV) 2026-06-16 DOI: arXiv:2606.16302

Explainable Flood Segmentation on Sentinel-1 SAR Imagery: A Comparative Study of CNN and Transformer Architectures

作者:

Arundhuti Banerjee ↗David Daou ↗

Rapid and accurate flood prediction is essential for disaster response and mitigation planning. Synthetic Aperture Radar (SAR) sensors in satellites are well-suited for this purpose because they operate independently of weather and daylight conditions. Although SAR-based data enable all-weather flood monitoring, distinguishing flooded land from permanent water remains a significant challenge, particularly when flooding is defined strictly as inundated land. This study provides a comprehensive comparison of convolutional neural network (CNN) and vision transformer architectures for multi-class flood segmentation using Sentinel-1 SAR imagery, specifically trained to separate flooded land from permanent water bodies and land. Three state-of-the-art (SOTA)CNN-based models, U-Net, U-Net++, and DeepLabV3 with ResNet-34 backbone, and three SegFormer variants (b0,b1,b2) were evaluated in two benchmark datasets, the ETCI NASA dataset and SenFloods11, using scene-based data splits to ensure a realistic assessment of spatial generalization. The results demonstrate that SegFormer-b2 significantly outperforms the U-Net baseline on the ETCI dataset (higher flood IoU across all 7 test scenes in the Wilcoxon signed-rank test), while after fine-tuning on Sen1Floods11, the advantage narrows to within the range of scene variability and is concentrated in spatially fragmented flood events. The study includes both qualitative and quantitative explainability techniques to visually comprehend model decisions and systematically assess prediction reliability. Qualitative analysis reveals that SegFormer-b2 produces more spatially coherent Grad-CAM activations focused on flood-relevant features, while U-Net generates more informative uncertainty estimates along flood boundaries.

阅读与讨论 → 访问原文 →

13.

arXiv (CS.AI) 2026-06-15 DOI: arXiv:2606.14240

AFFORDANCE20Q: Evaluating Affordance Reasoning from Physical Properties

作者:

Yifan Jiang ↗Meige Yang ↗Zitong Li ↗Jay Pujara ↗

arXiv:2606.14240v1 Announce Type: new Abstract: Affordance reasoning, the inference of an object's action possibilities from its physical properties (e.g., shape and material), is fundamental to human physical understanding and increasingly critical for Large Language Models (LLMs). However, existing affordance benchmarks largely expose explicit object identities in the evaluation setup, allowing models to rely on memorized object-affordance mappings rather than reasoning over physical properties. To address this gap, we introduce Affordance20Q, a novel affordance reasoning benchmark formulated as a 20-Questions game without exposing the object's identity. In each game, the model identifies a hidden object's affordance from a candidate set by asking yes/no questions about its physical properties. Affordance20Q comprises 1,009 games over 454 objects and 59 affordances, all manually filtered, refined, and annotated. We conduct comprehensive experiments with 15 state-of-the-art LLMs and find a substantial gap (~20 points) compared to human performance. A KL-based information-gain (IG) analysis further shows that models fail to ask discriminating questions as the game progresses. To close the gap, we develop KB-Anchored Rule Induction (KARI), a pipeline based on LLMs that generates affordance rules grounded in evidence from knowledge bases (KBs). KARI improves open-source LLMs by up to 15.2 points, while the limited coverage of KBs hinders further gains. We release all our code and data at https://github.com/1171-jpg/Affordance20Q.git

阅读与讨论 → 访问原文 →

14.

arXiv (CS.CV) 2026-06-17 DOI: arXiv:2606.17504

Two-Stage Fine-Tuning of ResNet50 for High-Sensitivity Melanoma Detection on Dermoscopic Images

作者:

Aryan Bhagat ↗

Melanoma is the most dangerous form of skin cancer with five-year survival rates exceeding 99% when detected early but falling sharply once the disease spreads. This paper proposes and evaluates a two-stage fine-tuning approach for ResNet50 applied to binary melanoma classification on dermoscopic images. The core challenges addressed are class imbalance and suboptimal transfer learning from single-stage fine-tuning. After stratified train/validation/test splitting, random oversampling was applied exclusively to the training set to achieve a 1:1 class balance. Stage 1 trained only the classification head with the ResNet50 base frozen, while Stage 2 fine-tuned all layers jointly at a low learning rate of 1e-5 to prevent catastrophic forgetting of learned visual features. On an independent test set of 3,826 images, the model achieved an AUC-ROC of 0.9559, accuracy of 88.34%, sensitivity of 87.56%, specificity of 89.13%, and F1-score of 88.29%. An ablation study confirms the two-stage protocol significantly outperforms single-stage fine-tuning, with sensitivity gains of over 4%. Grad-CAM visualizations demonstrate correct lesion localization. A fully deployable Streamlit detection application is provided alongside all training code.

阅读与讨论 → 访问原文 →

15.

bioRxiv (Bioinfo) 2026-06-14 DOI: HASH:0e30982991303c9a6c28f5f94d4aa2a6

Structural Analysis of Prostate Cancer N-Glycans Using Graph-Based Structural Metrics

作者:

Kalyanthaya ↗Gallegos ↗Chow ↗Pavletic ↗Diaz Fernandez ↗A. B ↗Kilcoyne ↗Joshi ↗Nguyen ↗D. H ↗

The N-linked glycans are structurally complex carbohydrate modifications that regulate protein folding, immune recognition, and cellular signaling, and their expression is extensively remodeled during cancer progression, making them promising biomarkers. In this study, prostate cancer-associated N-glycans from a range of relevant peer-reviewed studies were curated and digitized to develop a versatile computational framework that quantitatively encodes their spatial complexity across diverse biological systems. We invented two indices – the Distance & Connectivity Index (DCI) and the Position & Composition Index (PCI) – to capture the spatial information in N-glycans as layered architectures, enabling calculation of residue-level path lengths, branching structure, and compositional diversity. DCI summarizes glycan structure as both a scalar and matrix representation, while PCI does the same but also captures monosaccharide diversity, linkage heterogeneity, and cross-layer branching features. These metrics were computed with GlycoAssessor, an open-source platform that extracts information for the DCI and PCI from glycans drawn via Symbol Nomenclature for Glycans (SNFG) notation. Principal Component Analysis (PCA) was applied to evaluate whether glycans from prostate cancer tissues cluster distinctly in a disease-relevant manner. Results show that the spatial information in N-glycans: (1) increased in a multi-dimensional, non-linear manner, (2) objectively segregated structural themes, (3) could function as a potential prostate cancer biomarker that is distinct from mass-to-charge ratio and relative abundance, and (4) could objectively quantify novel subtype classifications of glycans associated with disease states and progression.

阅读与讨论 → 访问原文 →

16.

arXiv (CS.CL) 2026-06-16 DOI: arXiv:2606.14867

Evaluating the Robustness of Proof Autoformalization in Lean 4

作者:

Zhengtao Gui ↗Sheng Yang ↗Zhouxing Shi ↗

Proof autoformalization aims to translate a mathematical informal proof written in natural language into a formal proof in a formal language such as Lean~4. Several works have developed LLM-based models for proof autoformalization. However, existing evaluations have typically focused on translating well-formed informal proofs from curated datasets. We argue that a robust proof autoformalizer must remain faithful even for informal proofs that diverge from these idealized ones, and we present the first study on the robustness of proof autoformalization models. We formulate two categories of perturbations and evaluate robustness under each: a global perturbation paraphrases the informal proof in a different style, under which the formalization should remain consistent; a local perturbation alters a value, symbol, or proof step, possibly in a counterfactual way, and a robust formalization should faithfully reflect the perturbation rather than reverting to the original one or inferring a different one on its own. We build a benchmark with both perturbations on miniF2F and MATH-500, and automatically measure how stable a proof autoformalization's correctness is under global perturbations and how faithfully its output reflects local perturbations. We evaluate seven recent models, all of which are sensitive to global perturbations and mostly fail to remain faithful under local perturbations. Code and data are available via https://github.com/ucr-rai/robust-proof-autoformalization.

阅读与讨论 → 访问原文 →

17.

medRxiv (Medicine) 2026-06-15 DOI: HASH:61b067936df80506be2eb1f95aa0cb48

Artificial Intelligence-Based Detection of Airway Mucus Plugs on CT and Associations With Clinical Outcomes in COPDGene

作者:

Oyer ↗Namvar ↗Hoff ↗B. A ↗Bosma ↗Labaki ↗W. L ↗Kazerooni ↗E. A ↗Martinez ↗F. J ↗Hatt ↗…

RATIONALE: Airway mucus plugging is a clinically relevant manifestation of airway pathology in chronic obstructive pulmonary disease (COPD) and is associated with increased mortality even in early disease; however, visual computed tomography (CT) assessment is subjective and labor intensive. OBJECTIVES: To develop an AI-based quantitative CT method for automated detection of airway mucus plugging and evaluate associations with physiologic impairment and clinical outcomes. METHODS: Inspiratory CT scans from 8,971 COPDGene Phase 1 (GOLD 0-4 and PRISm) participants were analyzed. An AI-based framework combining 3D airway segmentation discontinuities and convolutional neural network classification identified mucus plug obstructions, yielding mucus plug burden (total plug count). Associations with outcomes were evaluated using covariate-adjusted models. MEASUREMENTS AND MAIN RESULTS : Higher mucus plug burden was associated with lower post-bronchodilator FEV % predicted ({rho} = -0.41; P < 0.001), greater air trapping (LAA < -856 HU; {rho} = 0.33; P < 0.001), worse health status (SGRQ; {rho} = 0.31; P < 0.001), and shorter 6-minute walk distance ({rho} = -0.26; P < 0.001). Among GOLD 1-4 participants, mucus plug presence was independently associated with increased all-cause mortality (adjusted hazard ratio, 1.28; P < 0.005) and exacerbation frequency (adjusted incidence rate ratio, 1.32; P < 0.005). Plug presence was also associated with increased respiratory mortality across GOLD categories and cardiovascular mortality in GOLD 1-2. CONCLUSIONS: AI-based quantitative CT assessment of airway mucus plugging provides a scalable, reproducible measure associated with physiologic impairment and adverse outcomes in COPD, supporting its role in risk stratification and future therapeutic studies.

阅读与讨论 → 访问原文 →

18.

arXiv (CS.CL) 2026-06-17 DOI: arXiv:2606.18124

Unintended Effects of Geographic Conditioning in Large Language Models

作者:

Naz Col ↗David M. Chan ↗

Modern conversational AI systems frequently rely on user metadata to localize responses, yet the unintended regional biases introduced by this hidden context remain poorly understood. In this work, we evaluate location leakage: the phenomenon where a model generates geographic references despite receiving a geographically neutral user prompt. Across both creative writing and open-ended Q&A prompts, even state-of-the-art LLMs systematically favor region-specific outputs when exposed to location metadata, with leakage spiking by up to 793 times above baseline (e.g., from 0.04% to 31.7% for Llama 3.1-8B, and 21.3% and 8.8% for Qwen3-8B and Claude Sonnet 4.6, respectively). Our analysis further shows a novel structural conditioning effect: replacing the injected location with the placeholder "Unknown" still elevates leakage by up to 72 times above baseline, demonstrating that the user profile frame itself, independent of any geographic content, acts as a generative conditioning signal.

阅读与讨论 → 访问原文 →

19.

arXiv (CS.AI) 2026-06-17 DOI: arXiv:2606.17328

MemTrace: Probing What Final Accuracy Misses in Long-Term Memory

作者:

Xianxuan Long ↗Zhikai Chen ↗Shenglai Zeng ↗Shouren Wang ↗Kai Guo ↗Jiliang Tang ↗

arXiv:2606.17328v1 Announce Type: new Abstract: LLM agents increasingly maintain long-term memory of user facts across sessions. Yet such memory is usually evaluated by aggregating accuracy over question rows or episodes. Because this approach scores question rows independently, even when several questions probe the same fact, it cannot show how that fact behaves as conditions change. We introduce MemTrace, a benchmark whose unit of measurement is the knowledge point: a single typed fact about the user, rather than an individual question. MemTrace probes each fact along three controlled dimensions: memory age, defined by how many sessions ago the fact appeared in the history; question type, covering current state, earlier state, and trajectory of change; and evidence condition, covering present, missing, and contradicted-by-false-premise settings. Evaluating 13 memory-system configurations across four paradigms, we find that similar pooled accuracy hides different failures: recovering a fact's current and earlier states does not imply tracking how it changed, and safe abstention does not imply correcting a false premise. The dominant bottleneck is evidence use, not retrieval: when systems fail, the evidence was retrievable 10 times more often than it was missing. These results suggest that improving long-term memory requires better use of reachable evidence, not simply more storage or retrieval.

阅读与讨论 → 访问原文 →

20.

arXiv (CS.AI) 2026-06-18 DOI: arXiv:2606.18996

TRAP: Benchmark for Task-completion and Resistance to Active Privacy-extraction

作者:

Moon Ye-Bin ↗Nam Hyeon-Woo ↗Baek Seong-Eun ↗Yejin Yeo ↗Tae-Hyun Oh ↗

arXiv:2606.18996v1 Announce Type: cross Abstract: Agents are increasingly deployed in document-intensive workflows where sensitive private information is not an edge case but a routine input, e.g., an agent booking a flight needs passport numbers. In such settings, the agent must use private information to complete tasks accurately while never exposing it in its responses, because it cannot verify who is actually at the keyboard. These two obligations are in fundamental tension. A model capable enough to use private information for task completion can, by the same capability, be induced to reveal it. To evaluate the trade-off of task accuracy and privacy leakage, we introduce Task-completion and Resistance to Active Privacy-extraction (TRAP). Each scenario includes a document containing private information, a task query that requires the agent to invoke the correct tool using private fields, and an attack query that attempts to elicit the same information in natural language. Evaluating 22 models spanning frontier proprietary and open-source models at multiple scales, we find that all model families exhibit non-trivial leakage, and that instruction-following ability correlates with leakage rate. Existing prompt-based defenses reduce leakage but at significant cost to task accuracy. Prompt optimization fails to escape this trade-off. We demonstrate that this failure is not incidental. For any softmax-based model, no soft-constraint defense, e.g., prompt-based defenses, can jointly achieve high task success with zero leakage probability. Motivated by this impossibility result, we propose structural private field isolation, which replaces private fields with hash keys before they reach the model. This approach largely prevents leakage while keeping task accuracy.

阅读与讨论 → 访问原文 →

21.

arXiv (CS.LG) 2026-06-12 DOI: arXiv:2606.12488

A Stationary (and Therefore Compatible) Representation is All You Need

作者:

Niccol\`o Biondi ↗Federico Pernici ↗Simone Ricci ↗Alberto Del Bimbo ↗

arXiv:2606.12488v1 Announce Type: new Abstract: Learning compatible representations aims to learn feature representations that can be used interchangeably over time whenever a model undergoes updates. In this paper, we demonstrate that stationary representations learned by d-Simplex fixed classifiers imply compatibility as in its formal definition. This result establishes a foundation for future works and can be directly exploited in practical learning scenarios. We address the challenge of learning compatibility using $d$-Simplex fixed classifiers when the model is sequentially fine-tuned. Learning according to a d-Simplex fixed classifier with the cross-entropy loss aligns feature distributions at the first-order statistics. Consequently, it may not fully capture higher-order dependencies in the representation between model updates. To address this issue, we demonstrate that training the model using a $d$-Simplex fixed classifier through a convex combination of the cross-entropy loss and a contrastive loss not only captures higher-order dependencies, but is also equivalent to learning with the cross-entropy under the compatibility constraints. We confirm our findings with extensive experiments also considering a new scenario where a pre-trained model is sequentially fine-tuned and occasionally replaced with an improved model. We show that stationary representations enable uninterrupted retrieval services (without reprocessing gallery images) while improving performance during model updates and replacements, achieving state-of-the-art. Code at https://github.com/miccunifi/iamcl2r.

阅读与讨论 → 访问原文 →

22.

arXiv (CS.LG) 2026-06-11 DOI: arXiv:2606.11256

My Chemical Harness: Evolutionary Molecular Design over Synthetic Pathways with Large Language Model Agents

作者:

C\'esar Ojeda ↗Darius A. Faroughy ↗Maryam Karimi ↗Payam Zarrintaj ↗Mir Mehdi Seyedebrahimi ↗Mart\'in Carballo-Pacheco ↗

arXiv:2606.11256v1 Announce Type: cross Abstract: Designing molecules with target properties is most useful when candidate structures are accompanied by feasible synthetic routes. We introduce My Chemical Harness, a route-native evolutionary framework for goal-directed molecular design in which the search population consists of executable synthetic pathways rather than isolated molecular graphs. Each route is built from purchasable building blocks and reaction templates, executed by deterministic chemistry tools, and scored through task-specific molecular oracles. Large language models (LLMs) are used only as strategy controllers that select high-level preferences over route length, move type, reaction families, motifs, and exploration pressure, while local code performs route construction, validation, deduplication, scoring, selection, and memory updates. This separation lets the LLM guide exploration without allowing it to introduce hallucinated products or unsupported reaction steps. On a soluble epoxide hydrolase proxy task, our LLM agent improves over single pass LLM and deterministic controllers, reaching state-of-the-art performance across the sEH score, synthetic accessibility score, and AiZynthFinder success rate metrics. These results suggest that constrained LLM agents can play a significant role in molecular discovery without requiring training, fine-tuning, or dedicated generative models.

阅读与讨论 → 访问原文 →

23.

arXiv (CS.AI) 2026-06-18 DOI: arXiv:2606.18315

Ghost Attractor Networks: Basin-Structured Dynamical Decoders for Closed-Loop Sequential Generation

作者:

Tianyu Wang ↗Ying Wang ↗Zhihao Liu ↗Xi Vincent Wang ↗Lihui Wang ↗

arXiv:2606.18315v1 Announce Type: cross Abstract: Sequential output generation with large-scale Transformer and diffusion decoders pays a memory cost that grows with sequence length, plus iterative per-step computation. Replacing them with small feed-forward decoders restores efficiency but produces unstructured latent representations that limit closed-loop control: phase-conditioned action generation and cross-step latent carry-over both require a latent geometry with stable basins. This article proposes Ghost Attractor Networks, a theoretically derived dynamical decoder whose latent evolves under a learned potential with drift and produces a basin-attractor structure by construction. Three desiderata (multi-modality, decoder-level single-pass switching, and constant memory) motivate the potential-drift form, and mode transitions arise as saddle-node bifurcations with ghost-attractor escape. A hierarchical phase-space decomposition separates first-order basin convergence from second-order proprioceptive refinement. Empirically, a Ghost trained end-to-end with a behavioral-cloning and contrastive objective exhibits the predicted gradient-flow contraction in its potential, with the gradient norm decaying by 67 percent across five integration steps on 1430 held-out samples. Ghost is evaluated as a robotic action decoder. A 2.3-million-parameter Ghost matches the offline accuracy of a 1.07-billion-parameter Diffusion Transformer at 462 times fewer parameters and 32 times lower latency, and beats five alternative 2M-parameter decoders (MLP, Neural ODE, CVAE, Transformer, 1-step Diffusion) on offline mean squared error by 5.9 to 29 percent. On the LIBERO-10 closed-loop benchmark, phase conditioning on Ghost's basin-structured latent yields a 13.5 percentage-point success-rate gain over a feed-forward MLP baseline, and persistent-latent ensembling reaches a 95.7 percent final success rate.

阅读与讨论 → 访问原文 →

24.

medRxiv (Medicine) 2026-06-16 DOI: HASH:07c28f73792a6ef89ffabb4c0df917db

Deployment-readiness audit of calibration, clinical utility, and fairness in perioperative infection prediction

作者:

Guillen-Ramirez ↗Lucas ↗K. L ↗Wintsch ↗Y. M ↗Blatter ↗T. U ↗Triep ↗Endrich ↗Beldi ↗

Objective: Clinical risk scores intended to guide patient-level decisions can show strong average performance. However, predicted probabilities can be systematically too high or too low in specific subgroups even when overall performance is strong. We audited deployment readiness of a strong end-of-surgery postoperative infection model across clinically relevant subgroups and tested mitigation strategies in miscalibrated subgroups. Materials and Methods: We analyzed out-of-fold predictions for 10,719 surgical procedures at a Swiss tertiary hospital, with 504 postoperative bacterial infection events. Prespecified axes were recorded sex, age stratum, and an EHR-derived physiological-reserve proxy. Within subgroups and pairwise intersections, we evaluated discrimination, calibration, threshold-specific errors, and decision-curve net benefit at the prespecified operating threshold. We compared group-specific isotonic recalibration with Wasserstein-barycenter postprocessing and demonstrated portability in SUPPORT2. Results: Overall AUROC was 0.876. While sex-marginal discrimination was similar in women and men (0.878 vs 0.875), age and reserve stratification revealed deployment-readiness failures. Calibration-in-the-large ranged from -0.86 in frail patients to -2.47 in non-frail patients. At the 0.10 operating threshold, decision-curve net benefit was positive in frail patients but negative in pre-frail and non-frail patients. Isotonic recalibration corrected average physiological-reserve-stratified calibration without worsening Brier scores, whereas Wasserstein postprocessing worsened calibration in most procedure clusters. Discussion: Discrimination-only or sex-marginal evaluation would have missed subgroup failures with clinical-utility implications. Conclusion: Subgroup fairness audits for clinical deployment should jointly evaluate discrimination, calibration, and utility. We implemented the audit as the open-source isitfair framework for identifying deployment-relevant subgroup failures, comparing mitigation strategies, and generating structured reports.

阅读与讨论 → 访问原文 →

25.

arXiv (CS.LG) 2026-06-11 DOI: arXiv:2606.12299

Learning What to Say to Your VLA: Mostly Harmless Vision Language Action Model Steering

作者:

Hyun Joe Jeong ↗Gokul Swamy ↗Andrea Bajcsy ↗

arXiv:2606.12299v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models provide a natural language interface to robot control, but the mapping from language to behavior is often brittle and unintuitive: semantically similar instructions can induce drastically different behaviors, while some capabilities may not be elicitable through prompting alone. As a result, both human instructions and zero-shot language models can fail to reliably steer VLAs toward successful task execution. In this work, we propose a framework that interactively searches for language sequences that improve closed-loop VLA task performance, distills these sequences into a test-time language feedback policy (LFP), and learns an improvement head that predicts when language steering will improve performance. We conformalize this improvement head to prevent harmful steering interventions, where the LFP decreases task performance relative to the original instruction on out-of-distribution scenarios. Crucially, our approach operates on arbitrary frozen pre-trained VLAs, requiring neither access to the original training distribution nor fine-tuning of the underlying model. On seen environments, our conformalized LFP improves base VLA performance by 24.7% in simulation and 65.0% in hardware. On visual and semantic perturbations, our conformalized LFP has strong harmlessness guarantees, and produces recovery behaviors not observed with open-loop prompting.

阅读与讨论 → 访问原文 →

探索全球前沿学术脉络