探索全球前沿学术脉络

01.

medRxiv (Medicine) 2026-06-15 DOI: HASH:ca0df469f0e50db3973b6b49a4680a2e

Two Blood-based Endotypes Reveal Divergent Clinical Outcomes of Fibrotic Hypersensitivity Pneumonitis

作者:

Huang↗Ma↗S.-F↗Kim↗J. S↗Strickland↗Receveur↗B. A↗Bonham↗C. S↗Paul↗T. K↗…

Rationale: Fibrotic hypersensitivity pneumonitis (fHP) is an antigen-driven, life-threatening interstitial lung disease characterized by heterogeneous radiologic features, clinical outcomes, and treatment responses. Objectives: To identify blood-based fHP endotypes that inform mechanism, prognosis and therapeutic response. Methods: We performed integrative analyses of multi-compartment transcriptomic data derived from whole blood, peripheral blood mononuclear cells, bronchoalveolar lavage, and surgical lung biopsies, alongside circulating plasma proteomics. Multiple clustering algorithms were cross-compared to ensure robustness and reproducibility of endotypes identification. Immune cell composition was inferred using bulk RNA-seq deconvolution and annotated with BAL single-cell RNA-seq. Pathway activities were characterized using Gene Set Enrichment Analysis. Transplant-free survival (TFS) was evaluated for endotype and corticosteroid exposure by Kaplan-Meier methods, with hazard ratios analyzed using multivariable Cox proportional hazards models. Results: Two molecular endotypes, lymphocytic-associated (L-fHP) and non-lymphocytic-associated (N-fHP), were identified and validated. L-fHP showed enrichment of adaptive immune signaling and lymphocyte predominance, whereas N-fHP demonstrated myeloid-cell activation with neutrophil and macrophage predominance. Corticosteroid exposure was associated with worse TFS in L-fHP but not in N-fHP after adjusting for age, sex, and baseline pulmonary function. Compared to L-fHP, N-fHP had poorer baseline pulmonary function, faster 12-month FVC decline, and shorter TFS. N-fHP also exhibited elevated neutrophil-associated markers, including matrix metalloproteinase-9, across paired transcriptomic and proteomic datasets, supporting a neutrophil-driven, cross-compartment disease process. Conclusion: Multi-omic, multi-compartment analysis identifies two reproducible fHP endotypes with distinct clinical outcomes and corticosteroid responses, supporting a precision medicine approach beyond current clinical and radiologic classification.

阅读与讨论 → 访问原文 →

02.

arXiv (CS.CV) 2026-06-18 DOI: arXiv:2606.03827

Conditional Latent Diffusion Model with Fourier-based Motion Modelling for Virtual Population Synthesis

作者:

Shaokun Lan↗Haoran Dou↗Jinghan Huang↗Arezoo Zakeri↗Fengming Lin↗Zherui Zhou↗Jinming Duan↗Alejandro F. Frangi↗

In-silico trials of medical devices require the generation of virtual populations of anatomies. In cardiovascular applications, virtual anatomy is typically represented as a 3D+t mesh sampled from a generative model. However, most existing mesh generators focus on static anatomy, while sequence models often lack explicit periodicity. To this end, we propose 4D F-MeshLDM, a conditional generative framework comprising a convolutional mesh VAE to encode meshes, a structural latent space that parameterises motion using a truncated Fourier series, and a diffusion prior that learns the latent distribution over Fourier coefficient tokens. By conditioning the diffusion process on clinical covariates via affine modulation, we enable controllable synthesis. Sampling tokens and performing inverse Fourier synthesis yield cycle-consistent latent trajectories, which can be decoded into 3D+t cardiac mesh sequences. Experiments on 5,000 UK Biobank subjects demonstrate that 4D F-MeshLDM outperforms state-of-the-art baselines in anatomical fidelity and achieves near-zero cycle closure error. Furthermore, the generated cohorts accurately preserve clinical functional indices, highlighting the potential of our framework for reliable in-silico cardiac trials.

阅读与讨论 → 访问原文 →

03.

arXiv (CS.CV) 2026-06-16 DOI: arXiv:2606.15167

Variational Network with Wavelet-based UNET in Accelerated MRI Reconstruction from Under Sampled K-space Data

作者:

Yasir Arafat Prodhan↗Shaikh Anowarul Fattah↗

Fully sampled MRI requires dense k-space acquisition, leading to long scan times, reduced clinical throughput, and increased sensitivity to patient motion. Accelerated MRI addresses this by acquiring undersampled k-space data and reconstructing the missing information computationally. However, reconstruction from undersampled measurements is highly ill-posed and can introduce aliasing artifacts, noise amplification, and loss of anatomical detail. Although conventional parallel imaging and compressed sensing methods mitigate these issues, and deep learning methods have further improved reconstruction quality, preserving high-frequency structures under aggressive undersampling remains challenging. In this work, we propose a Variational Network with a Wavelet-based U-Net (W-UNet) for accelerated MRI reconstruction. The framework combines physics-guided iterative reconstruction with learnable multi-scale frequency representations. Standard pooling operations are replaced with Discrete Wavelet Transform and Inverse Wavelet Transform modules, enabling lossless downsampling while preserving low-frequency structure and high-frequency edge details. Integrated into the refinement and sensitivity map estimation stages, the proposed design improves artifact suppression, feature preservation, and reconstruction fidelity in both single-coil and multi-coil settings. Experiments on fastMRI knee and M4Raw brain datasets show state-of-the-art performance. Ablation studies further confirm the effectiveness of wavelet-based feature decomposition for accelerated MRI reconstruction.

阅读与讨论 → 访问原文 →

04.

arXiv (CS.LG) 2026-06-18 DOI: arXiv:2606.19117

Wasserstein Policy Learning for Distributional Outcomes

作者:

Yiyan Huang↗Cheuk Hang Leung↗Qi Wu↗Zhiheng Zhang↗

arXiv:2606.19117v1 Announce Type: cross Abstract: Offline policy learning has received growing attention in causal inference. The primary objective is to learn a policy (individualized treatment rule) as a mapping from covariates to treatment that maximizes the empirical welfare defined as the mean of scalar-valued potential outcomes. In this paper, we study offline policy learning with distribution-valued outcomes, where each potential outcome is a probability measure on $\mathbb{R}$ and the reward is defined through a utility functional applied to the Wasserstein barycenter of induced outcome distributions. We establish statistical guarantees for the policy learning framework based on both Inverse Probability Weighting (IPW) and Doubly Robust (DR) estimators. By handling the challenging uniform deviation over the product of the combinatorial policy class and the infinite-dimensional quantile domain, we prove that the finite-sample regret has leading dependence $\widetilde{\mathcal{O}}(\sqrt{\mathrm{N-dim}(\Pi)/N})$. In the one-dimensional Wasserstein setting and under the stated regularity conditions, the leading regret rate is still governed by the policy-class complexity. Moreover, we provide a minimax lower bound establishing the sharpness of the leading dependence on $N$ and $\mathrm{N-dim}(\Pi)$.

阅读与讨论 → 访问原文 →

05.

arXiv (CS.CL) 2026-06-11 DOI: arXiv:2606.11512

SAGE: Answer-Conditioned Uncertainty Targets for Verbal Uncertainty Alignment

作者:

Kaiwen Shi↗Zheyuan Zhang↗Yanfang Ye↗

Large language models increasingly express uncertainty through natural-language statements, yet these expressions often fail to reflect the model's sampled behavior. We study verbal uncertainty alignment as a distributional calibration problem: the appropriate uncertainty target for a prompt should be estimated from repeated model outputs rather than from an isolated response. However, group rollouts alone are insufficient, since the resulting target must provide a useful training signal. Existing targets only partially satisfy this requirement. We propose SAGE, Semantic-Answer Guided Entropy, a group-level uncertainty target that constructs an answer-conditioned uncertainty geometry over sampled responses. SAGE preserves categorical, numeric, and symbolic answer distinctions while maintaining a smooth and scale-preserving calibration signal. We further apply this target through Group-Uncertainty Preference Optimization, or GUPO, an uncertainty-channel training framework that supervises verbal uncertainty expressions rather than the full response. Experiments across factual, mathematical, and multiple-choice reasoning tasks show improved uncertainty ranking, lower calibration error, and reduced overconfidence.

阅读与讨论 → 访问原文 →

06.

arXiv (CS.CL) 2026-06-17 DOI: arXiv:2606.17250

Rethinking Groups in Critic-Free RLVR

作者:

Yihong Wu↗Liheng Ma↗Lingfeng Xiao↗Muzhi Li↗Xinyu Wang↗Yingxue Zhang↗Jian-Yun Nie↗

Reinforcement learning (RL) has become a central paradigm for post-training large language models. Existing critic-free RL methods typically generate a group of rollouts for the same question to estimate value baselines for advantage computation. However, this design suffers from data inefficiency, group synchronization barriers, and inflexibility with structured rollouts. In this work, we revisit the role of the ``group'' and show that its underlying function is not merely to estimate baselines but to prevent false penalties on negative samples. Building on this insight, we propose negative token filtering, a simple and effective strategy that enables stable single-rollout training. We apply it to two batch-level advantage methods, achieving comparable performance on reasoning tasks and stronger performance on agentic tasks relative to group-based RL techniques.

阅读与讨论 → 访问原文 →

07.

arXiv (CS.CL) 2026-06-19 DOI: arXiv:2606.19946

GEMS: Geometric Constraints Enable Multi-Semantic Superposition in LLMs

作者:

Yu Deng↗

Activation steering controls model behavior by modifying intermediate hidden states at inference time without retraining. Existing methods handle only single-direction injection; when multiple semantic directions are superposed without constraints, the model collapses. We show that this collapse decomposes into two independently acting sources: distributional deviation, where additive perturbations accumulate in norm across layers and drive activations outside the training distribution, and directional interference, where non-orthogonal semantic vectors mutually dampen when superposed. These two sources define the design constraints that any training-free multi-directional intervention must address. As one instantiation of these principles, we propose GEMS, a training-free method that maps each source to a corresponding geometric constraint: norm-preserving weighted superposition and targeted attention-pathway injection for distributional deviation, and real-time orthogonalization for directional interference. On GSM8K, injecting three concurrent non-mathematical directions preserves accuracy at 98% (baseline 92%), while unconstrained addition collapses to 4%; on Wikitext-2, the same injection incurs only 2.2% PPL increase. Component ablation isolates the causal role of each constraint, and layer-level probes confirm that orthogonalized signals survive the FFN pathway and reach the output distribution with semantic specificity. Qualitative steering effects transfer across architectures from 3B to 31B.

阅读与讨论 → 访问原文 →

08.

arXiv (CS.CV) 2026-06-15 DOI: arXiv:2606.13700

C-MambaPose: A Physics-Informed Complex Mamba Framework for Cross-Environment WiFi Human Pose Estimation

作者:

Phuc Nguyen H↗

Human pose estimation (HPE) utilizing wireless WiFi signals has emerged as a promising technology owing to its device-free nature, privacy preservation, and robustness against occlusion and poor lighting. However, existing methods often overlook the physical complex phase information of WiFi signals and fail to generalize across diverse environments due to severe domain shifts. In this paper, we present C-MambaPose, a physics-informed complex-valued Mamba-GraFormer hybrid framework for robust cross-environment WiFi-based 3D HPE. Our framework first sanitizes raw WiFi Channel State Information (CSI) phase errors and constructs a phase-preserving complex-valued representation. We then employ a Spatiotemporal Complex Mamba encoder with a dynamic selective receptive field to capture fine-grained phase dynamics. A cross-attention joint-query mapper maps the unstructured sequence tokens to human joints, which are decoded by a Graph Convolutional Network (GCN) to predict anatomically coherent 3D coordinates. Extensive evaluations on the MM-Fi dataset show that C-MambaPose achieves competitive or superior performance to state-of-the-art baselines across all settings, setting a new state-of-the-art specifically on the challenging cross-environment split, requiring only 3.78 M parameters-an 83.1\% reduction compared to GraphPose-Fi[chen2026graph] and an 85.7\% reduction compared to MetaFi++[zhou2023metafi++], while maintaining a comparable size to DT-Pose[chen2025towards] (which is only 18\% smaller) but achieving significantly superior performance without requiring any pretraining. Our code is publicly available at https://github.com/phucngvinuni/cmampose.git.

阅读与讨论 → 访问原文 →

09.

arXiv (CS.AI) 2026-06-17 DOI: arXiv:2606.17551

Reversal Q-Learning

作者:

Aditya Oberai↗Seohong Park↗Sergey Levine↗

arXiv:2606.17551v1 Announce Type: cross Abstract: Iterative generative modeling techniques, such as flow matching, provide powerful tools to model complex behaviors for effective offline reinforcement learning (RL). In this work, we propose a new off-policy RL algorithm that trains a flow policy based on prior data. Our idea starts from the "expanded" Markov decision process (MDP) framework, which treats individual flow refinement steps as separate actions in an MDP. To enable off-policy RL within this framework, we apply two techniques: we generate virtual on-policy trajectories (by "reversing" flows) to make this framework compatible with prior data, and we apply a bias-and-variance reduction technique to mitigate the curse of horizon in off-policy RL. We call the resulting algorithm Reversal Q-learning (RQL). RQL has several advantages over previous flow-based RL methods: it does not suffer from backpropagation through time, makes better use of the learned value function, and directly trains the full, expressive flow policy. Through our experiments on 50 challenging simulated robotic tasks, we show that RQL leads to the best average offline RL performance compared to state-of-the-art flow-based offline RL algorithms.

阅读与讨论 → 访问原文 →

10.

arXiv (CS.CL) 2026-06-19 DOI: arXiv:2606.19698

What sentiment analysis can't see: Measuring whether customers were helped, and what went wrong, across 70,000 support conversations

作者:

Jason Potteiger↗

Most companies read their customer support data at scale using sentiment analysis, which measures how customers sound rather than whether they were satisfied with the result. We tested a richer alternative on 70,450 support conversations from a leading online fundraising platform: alongside tone, we used GPT-5.4 to estimate each customer's satisfaction and to flag whether they reported a concrete problem, then validated all three readings against the 1-to-5 ratings customers left on the conversations they rated. The satisfaction estimate tracked those ratings far better than sentiment did, correlating at 0.47 against 0.36 and flagging unhappy customers with far fewer false alarms. The structured read also sees what sentiment cannot: tone and satisfaction disagree in 44% of conversations, a single "Neutral" label hides everything from quietly satisfied customers to ones who quietly gave up, and the largest group of all is "tolerated friction," customers who are satisfied but still reporting a fixable problem, a standing issue that no sentiment-based dashboard can surface. The broader finding is that LLM-based annotation can capture far more than the tonality of a customer's language, offering strong potential for new business metrics grounded instead in the customer's state (whether they were satisfied) and the cause of their problem extracted directly from the raw textual data of interactions and feedback.

阅读与讨论 → 访问原文 →

11.

arXiv (CS.CV) 2026-06-16 DOI: arXiv:2606.15287

G2IA: Geometry-Guided Instance-Aware Retrieval and Refinement for Cross-Modal Place Recognition

作者:

Xianyun Jiao↗Jingyi Xu↗Zhongmiao Yan↗Xieyuanli Chen↗Lin Pei↗

Cross-modal place recognition (CMPR) enables camera-only robots to localize against pre-built LiDAR maps in autonomous navigation scenarios. This image-to-point-cloud setting is challenged by two coupled ambiguities: the modality gap between perspective RGB appearance and sparse metric geometry, and perceptual aliasing among urban places with similar roads, facades, intersections, and object arrangements. Instead of treating CMPR as a single global descriptor matching problem, we argue that reliable retrieval requires both geometry-aware representation alignment and fine-grained candidate verification. In this paper, we propose G2IA, a geometry-guided instance-aware framework for image-to-point-cloud place recognition. In the retrieval stage, visual geometry priors from VGGT and instance features are integrated to construct place descriptors that are more compatible with LiDAR-derived map representations. In the refinement stage, the retrieved candidates are re-ranked by explicitly verifying whether local instance shapes and their relative spatial layouts are consistent across modalities. Experiments on public benchmarks demonstrate that G2IA consistently improves image-to-point-cloud place recognition under different localization thresholds, and exhibits strong cross-dataset generalization.

阅读与讨论 → 访问原文 →

12.

arXiv (CS.AI) 2026-06-18 DOI: arXiv:2605.27729

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

作者:

Dongping Liu↗Aoyu Zhang↗Luyao Zhang↗

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

阅读与讨论 → 访问原文 →

13.

arXiv (CS.AI) 2026-06-11 DOI: arXiv:2505.17623

\texttt{Range-Arithmetic}: Verifiable Deep Learning Inference on an Untrusted Party

作者:

Ali Rahimi↗Babak H. Khalaj↗Mohammad Ali Maddah-Ali↗

arXiv:2505.17623v2 Announce Type: replace-cross Abstract: Verifiable computing (VC) has gained prominence in decentralized machine learning systems, where resource-intensive tasks like deep neural network (DNN) inference are offloaded to external participants due to blockchain limitations. This creates a need to verify the correctness of outsourced computations without re-execution. We propose \texttt{Range-Arithmetic}, a novel framework for efficient and verifiable DNN inference that transforms non-arithmetic operations, such as rounding after fixed-point matrix multiplication and ReLU, into arithmetic steps verifiable using sum-check protocols and concatenated range proofs. Our approach avoids the complexity of Boolean encoding, high-degree polynomials, and large lookup tables while remaining compatible with finite-field-based proof systems. Experimental results show that our method not only matches the performance of existing approaches, but also reduces the computational cost of verifying the results, the computational effort required from the untrusted party performing the DNN inference, and the communication overhead between the two sides.

阅读与讨论 → 访问原文 →

14.

arXiv (CS.CL) 2026-06-18 DOI: arXiv:2606.01249

Trust Region On-Policy Distillation

作者:

Xingrun Xing↗Haoqing Wang↗Boyan Gao↗Ziheng Li↗Yehui Tang↗

On-Policy Distillation (OPD) is a fundamental technique for efficient post-training of large language models (LLMs), with broad applications in agent learning, multi-task enhancement, and model compression. However, OPD training becomes unstable when the teacher and student distributions differ substantially, as teacher supervision on student-generated tokens may yield unreliable policy gradients and even cause optimization failure. This work addresses reliable on-policy token-level supervision through credit assignment strategies, and proposes Trust Region On-Policy Distillation, TrOPD. It features the following characteristics: 1) Trust-Region On-Policy Learning: TrOPD performs OPD only in regions where the teacher provides reliable supervision, mitigating the optimization difficulty of the K1 reverse-KL estimator under distribution mismatch. 2) Outlier Estimation: For outlier regions, we explore gradient clipping, masking, and forward-KL estimation to reduce the adverse effects of unreliable supervision. 3) Off-Policy Guidance: The student continues generation from teacher prefixes and uses forward KL to imitate off-policy guidance, encouraging on-policy exploration toward reliable regions. Experiments show that TrOPD consistently outperforms SoTA OPD baselines, including OPD, EOPD, and REOPOLD, across mathematical reasoning, code generation, and general-domain benchmarks.

阅读与讨论 → 访问原文 →

15.

arXiv (CS.CL) 2026-06-16 DOI: arXiv:2606.16140

VibeThinker-3B: Exploring the Frontier of Verifiable Reasoning in Small Language Models

作者:

Sen Xu↗Shixi Liu↗Wei Wang↗Jixin Min↗Yingwei Dai↗Zhibin Yin↗Yirong Chen↗Xin Zhou↗Junlin Zhang↗

This technical report introduces VibeThinker-3B, a compact dense model with 3B parameters developed to investigate how far verifiable reasoning can be pushed within a strictly small-model regime. Building upon the Spectrum-to-Signal post-training paradigm, we systematically enhance the model through an optimized pipeline that includes curriculum-based supervised fine-tuning, multi-domain reinforcement learning, and offline self-distillation. Experimental evaluations demonstrate that VibeThinker-3B achieves frontier-level performance on highly demanding verifiable tasks. Specifically, it attains a score of 94.3 on AIME26 (improving to 97.1 with claim-level test-time scaling), an 80.2 Pass@1 on LiveCodeBench v6, and exhibits strong out-of-distribution generalization with a 96.1\% acceptance rate on recent unseen LeetCode contests. This effectively places it in the performance band of first-tier reasoning systems, matching or exceeding flagship models that are orders of magnitude larger, such as DeepSeek V3.2, GLM-5, and Gemini 3 Pro. Furthermore, a score of 93.4 on IFEval confirms that this extreme reasoning enhancement does not compromise strict instruction controllability. Extending our previous 1.5B work, these findings motivate the Parametric Compression-Coverage Hypothesis, which views verifiable reasoning as compressible into compact reasoning cores, while open-domain knowledge and general-purpose competence require broad parameter coverage over facts, concepts, and long-tail scenarios. This perspective suggests that compact models are not merely deployment-efficient substitutes, but a complementary path toward frontier-level performance in parameter-dense capability regimes.

阅读与讨论 → 访问原文 →

16.

arXiv (CS.CL) 2026-06-18 DOI: arXiv:2503.04989

Application of integrated gradients explainability to sociopsychological semantic markers

作者:

Ali Aghababaei↗Jan Nikadon↗Magdalena Formanowicz↗Maria Laura Bettinsoli↗Carmen Cervone↗Caterina Suitner↗Tomaso Erseghe↗

Classification of textual data in terms of sentiment, or more nuanced sociopsychological markers (e.g., agency), is now a popular approach commonly applied at the sentence level. In this paper, we exploit the integrated gradient (IG) method to capture the classification output at the word level, revealing which words actually contribute to the classification process. This approach improves explainability and provides in-depth insights into the text. We focus on sociopsychological markers beyond sentiment and investigate how to effectively train IG in agency, one of the very few markers for which a verified deep learning classifier, BERTAgent, is currently available. Performance and system parameters are carefully tested, alternatives to the IG approach are evaluated, and the usefulness of the result is verified in a relevant application scenario. The method is also applied in a scenario where only a small labeled dataset is available, with the aim of exploiting IG to identify the salient words that contribute to building the different classes that relate to relevant sociopsychological markers. To achieve this, an uncommon training procedure that encourages overfitting is employed to enhance the distinctiveness of each class. The results are analyzed through the lens of social psychology, offering valuable insights.

阅读与讨论 → 访问原文 →

17.

arXiv (CS.AI) 2026-06-17 DOI: arXiv:2606.17087

ZIVARI-TLBO: A Zero-Cost Inter-Group Evaluated-Elite Relay Mechanism for Teaching-Learning-Based Optimization

作者:

Pezhman Zivari↗

arXiv:2606.17087v1 Announce Type: cross Abstract: ZIVARI-TLBO is a grouped Teaching-Learning-Based Optimization (TLBO) method that augments an existing population-state controller with a fixed inter-group evaluated-elite relay. At each scheduled event, every group offers its already evaluated elite to the next group in a fixed ring; the elite replaces the receiver's worst eligible learner only when its stored objective value is better. Because the exact relay copies an already evaluated solution and its stored fitness, it requires no additional objective-function calls. The frozen gts-v4-cm-fixed implementation is evaluated under equal 10,000-evaluation budgets on eight classical functions at dimensions 10, 30, 50, and 100, with 30 matched seeds, and on five constrained engineering problems. A direct ablation against the same grouped landscape-aware controller without relay records 728/11/221 wins/ties/losses and a rank-biserial effect size of 0.624 across dimensions. In an eight-method multidimensional comparison, WOA obtains the best average rank (2.914) and ZIVARI-TLBO ranks second (3.382); ZIVARI-TLBO significantly outperforms TLBO, MCTLBO, DE, PSO, and GWO, loses significantly to WOA, and is not significantly different from HHO after Holm adjustment. Feasibility-aware engineering results are mixed and sensitive to the current static-penalty formulation. The evidence supports a scoped relay contribution and budget-consistent information-sharing mechanism, but not universal state-of-the-art, global-convergence, engineering-dominance, or CEC superiority claims.

阅读与讨论 → 访问原文 →

18.

arXiv (CS.CL) 2026-06-16 DOI: arXiv:2606.16497

daVinci-kernel: Co-Evolving Skill Selection, Summarization, and Utilization via RL for GPU Kernel Optimization

作者:

Dayuan Fu↗Mohan Jiang↗Tongyu Wang↗Dian Yang↗Jiarui Hu↗Liming Liu↗Jinlong Hou↗Pengfei Li↗

GPU kernel optimization represents a paradigm where functional correctness is assumed and execution efficiency is the objective. We present daVinci-kernel, a reinforcement learning framework that couples skill discovery with skill exploitation through a dynamically evolving skill library. daVinci-kernel jointly trains three agents sharing one LLM backbone: a Skill Selection Agent that retrieves relevant techniques via BM25 and LLM reranking, a Policy Agent that generates multi-turn CUDA/Triton kernels conditioned on selected skills, and a Skill Summary Agent that distills successful rollouts into reusable skills. Candidate skills are added only after execution-based verification confirms reproducible speedups. All three agents share a single LLM backbone, are initialized via a structured SFT cold start on diversity-filtered data, and are then jointly optimized end-to-end with multi-turn REINFORCE and per-agent advantage estimation. On KernelBench, daVinci-kernel-14B achieves 37.2%, 70.6%, and 32.2% on Level 1, Level 2, and Level 3 under the Fast$_1$ threshold, outperforming the strongest prior RL-trained model, Dr.Kernel-14B.

阅读与讨论 → 访问原文 →

19.

arXiv (CS.CL) 2026-06-11 DOI: arXiv:2606.12087

FORT-Searcher: Synthesizing Shortcut-Resistant Search Tasks for Training Deep Search Agents

作者:

Jia Deng↗Yimeng Chen↗Xiaoqing Xiang↗Ziyang Zeng↗Shuo Tang↗Wayne Xin Zhao↗Feng Chang↗Chuan Hao↗Yuan Wei↗Ran Tao↗Bryan Dai↗Ji-Rong Wen↗…

Training deep search agents requires verifiable questions whose answers remain unavailable until sufficient evidence has been acquired through search. Existing synthesis methods often increase apparent difficulty by enriching graph structures, but structural complexity alone does not guarantee realized search difficulty: the intended search process can collapse through a cheaper identifying route. We formalize this gap with a shortcut-aware difficulty framework and identify four actionable shortcut risks: evidence co-coverage, single-clue selectivity, exposed constants, and prior-knowledge binding. To diagnose their realized effects, we use trajectory signatures including solving cost, answer hit time, and prior-shortcut rate. Guided by this framework, we introduce FORT, a Framework of Shortcut-Resistant Training-Data Synthesis. FORT constructs shortcut-resistant training data by controlling shortcut risks across entity selection, evidence graph construction, question formulation, and adversarial refinement. Experiments show that FORT induces longer pre-answer search and fewer shortcut patterns than existing open-source deep search datasets. Using the resulting trajectories, we train FORT-Searcher with supervised fine-tuning (SFT) only, and it achieves the best overall performance among comparable-size open-source search agents on challenging deep search benchmarks. Relevant resources will be made available at https://github.com/RUCAIBox/FORT-Searcher.

阅读与讨论 → 访问原文 →

20.

arXiv (CS.CL) 2026-06-11 DOI: arXiv:2605.12288

TokenRatio: Principled Token-Level Preference Optimization via Ratio Matching

作者:

Truong Nguyen↗Tien-Phat Nguyen↗Linh Ngo Van↗Duy Minh Ho Nguyen↗Khoa Doan↗Trung Le↗

Direct Preference Optimization (DPO) is a widely used RL-free method for aligning language models from pairwise preferences, but it models preferences over full sequences even though generation is driven by per-token decisions. Existing token-level extensions typically decompose a sequence-level Bradley-Terry objective across timesteps, leaving per-prefix (state-wise) optimality implicit. We study how to recover token-level preference optimality using only standard sequence-level pairwise comparisons. We introduce Token-level Bregman Preference Optimization (TBPO), which posits a token-level Bradley-Terry preference model over next-token actions conditioned on the prefix, and derive a Bregman-divergence density-ratio matching objective that generalizes the logistic/DPO loss while preserving the optimal policy induced by the token-level model and maintaining DPO-like simplicity. We introduce two instantiations: TBPO-Q, which explicitly learns a lightweight state baseline, and TBPO-A, which removes the baseline through advantage normalization. Across instruction following, helpfulness/harmlessness, and summarization benchmarks, TBPO improves alignment quality and training stability and increases output diversity relative to strong sequence-level and token-level baselines.

阅读与讨论 → 访问原文 →

21.

arXiv (CS.LG) 2026-06-16 DOI: arXiv:2606.14763

Bayesian Optimization for Learning Nonlinear MPC in Autonomous Agent Navigation

作者:

Lorenzo Ortolani↗Gabriel Voss↗Gabriele Beltrami↗Francesco Dorati↗Tommaso Felice Banfi↗

arXiv:2606.14763v1 Announce Type: cross Abstract: Real-time autonomous navigation in dynamic, unknown environments remains a fundamental challenge for mobile robotics. We propose a map-free framework that tightly integrates reactive rolling-horizon planning with nonlinear Model Predictive Control (MPC). At each control cycle, a LiDAR-based Gaussian occupancy representation is constructed and used to generate collision-free trajectories via A* search, which are then tracked by a CasADi/IPOPT MPC formulation incorporating a smooth sigmoid obstacle barrier. To improve robustness to parameter sensitivity, we adopt an offline Bayesian optimization scheme based on Tree-structured Parzen Estimators (TPE), which identifies near-optimal controller parameters with respect to a composite navigation objective. In addition, a Gaussian Process surrogate is used to analyze parameter sensitivity and provide insight into the optimization landscape. The proposed framework is robot-agnostic and is evaluated on the Unitree Go2 quadruped in simulation using Gazebo, followed by deployment on the physical robot. Experimental results show that parameters tuned in simulation transfer effectively to hardware, maintaining comparable performance without additional tuning. The full system achieves up to a 90.0\% navigation success rate when deployed, along with a 38.9\% average improvement in the evaluation metrics across simulated environments.

阅读与讨论 → 访问原文 →

22.

arXiv (CS.CV) 2026-06-17 DOI: arXiv:2604.00605

Fluently Lying: Adversarial Robustness Can Be Substrate-Dependent

作者:

Daye Kang↗Hyeongboo Baek↗

The primary tools used to monitor and defend object detectors under adversarial attack assume that when accuracy degrades, detection count drops in tandem. This coupling was assumed, not measured. We report a counterexample observed on a single model: under standard PGD, EMS-YOLO, a spiking neural network (SNN) object detector, retains more than 70% of its detections while mAP collapses from 0.528 to 0.042. We term this count-preserving accuracy collapse Quality Corruption (QC), to distinguish it from the suppression that dominates untargeted evaluation. Across four SNN architectures and two threat models (l-infinity and l-2), QC appears only in one of the four detectors tested (EMS-YOLO). On this model, all five standard defense components fail to detect or mitigate QC, suggesting the defense ecosystem may rely on a shared assumption calibrated on a single substrate. These results provide, to our knowledge, the first evidence that adversarial failure modes can be substrate-dependent.

阅读与讨论 → 访问原文 →

23.

arXiv (quant-ph) 2026-06-16 DOI: arXiv:2606.15605

Dressed Floquet scars from protected zero modes in a Rydberg chain

作者:

Saptadip Roy↗Bhaskar Mukherjee↗K. Sengupta↗Arnab Sen↗

arXiv:2606.15605v1 Announce Type: cross Abstract: In this Letter, we present an approximate analytic construction of two zero quasienergy quantum many-body scars in a periodically driven model of Rydberg atoms on a ring, which persist over a range of driving amplitudes and frequencies for finite sizes. An index theorem protects an exponentially large number (in system size) of exact zero energy modes of the Floquet Hamiltonian in this setting. Unlike most of these zero modes which continuously change with drive parameters, these two quantum many-body scars retain the memory of particular states. They can be expressed as {\it dressed versions} of two contrasting states, the Rydberg vacuum and a unitarily rotated variant of a volume-law scar [Ivanov and Motrunich, Phys. Rev. Lett. {\bf 134}, 050403 (2025)], respectively. We provide an analytic understanding of their existence using a Floquet perturbation theory and show their resilience beyond the perturbative regime using exact diagonalization in finite systems. Our study provides insight into the structure of protected zero modes in interacting Floquet settings.

阅读与讨论 → 访问原文 →

24.

arXiv (CS.AI) 2026-06-17 DOI: arXiv:2605.05172

When Life Gives You BC, Make Q-functions: Extracting Q-values from Behavior Cloning for On-Robot Reinforcement Learning

作者:

Lakshita Dodeja↗Ondrej Biza↗Shivam Vats↗Stephen Hart↗Stefanie Tellex↗Robin Walters↗Karl Schmeckpeper↗Thomas Weng↗

arXiv:2605.05172v2 Announce Type: replace-cross Abstract: Behavior Cloning (BC) has emerged as a highly effective paradigm for robot learning. However, BC lacks a self-guided mechanism for online improvement after demonstrations have been collected. Existing offline-to-online learning methods often cause policies to replace previously learned good actions due to a distribution mismatch between offline data and online learning. In this work, we propose Q2RL, Q-Estimation and Q-Gating from BC for Reinforcement Learning, an algorithm for efficient offline-to-online learning. Our method consists of two parts: (1) Q-Estimation extracts a Q-function from a BC policy using a few interaction steps with the environment, followed by online RL with (2) Q-Gating, which switches between BC and RL policy actions based on their respective Q-values to collect samples for RL policy training. Across manipulation tasks from D4RL and robomimic benchmarks, Q2RL outperforms SOTA offline-to-online learning baselines on success rate and time to convergence. Q2RL is efficient enough to be applied in an on-robot RL setting, learning robust policies for contact-rich and high precision manipulation tasks such as pipe assembly and kitting, in 1-2 hours of online interaction, achieving success rates of up to 100% and up to 3.75x improvement against the original BC policy. Code and video are available at https://pages.rai-inst.com/q2rl_website/

阅读与讨论 → 访问原文 →

25.

arXiv (CS.CV) 2026-06-12 DOI: arXiv:2606.13303

DuET: Dual Expert Trajectories for Diffusion Image Editing

作者:

Lidia Troeshestova↗Alexander Ustyuzhanin↗Sergey Kastryulin↗

Recent diffusion editors perform diverse instruction-based edits while conditioning on the source image at every denoising step. Yet persistent source-image conditioning can limit how fully an edit is executed and how natural the result appears, especially when the target scene diverges substantially from the input. We introduce DuET (Dual Expert Trajectories), a training-free inference method that temporarily relaxes source-image conditioning by transitioning through a text-to-image phase before returning to edit mode, allowing the denoising trajectory to move toward the target distribution while retaining the structural benefits of image-conditioned editing. Without modifying model weights or increasing sampling cost, DuET consistently improves instruction relevance, semantic fidelity, and perceptual quality across diverse models and benchmarks. In some cases, these gains come with a modest reduction in source-image preservation, revealing a predictable trade-off between source preservation and edit fidelity.

阅读与讨论 → 访问原文 →