探索全球前沿学术脉络

01.

arXiv (CS.AI) 2026-06-12 DOI: arXiv:2606.13097

Functional Cache Grafting: Robust and Rapid Code-Policy Synthesis for Embodied Agents

作者:

Saehun Chun↗Wonje Choi↗Sera Choi↗Sanghyun Ahn↗Honguk Woo↗

arXiv:2606.13097v1 Announce Type: cross Abstract: Code-writing large language models (CodeLLMs) generate executable code policies for embodied agents by translating natural language goals and environmental constraints into structured control programs. However, policy generation in open-domain embodied environments suffers from two fundamental limitations: (i) delayed decoding caused by repetitive prefill computation over long prompts, and (ii) limited robustness due to fully generative decoding, which often produces API mismatches, missing safety guards, and unstable control logic. To address these limitations, we present FCGraft, a Functional Cache Grafting framework. FCGraft maintains a library of function-level validated code skeletons and their associated prompt-level Transformer key-value (KV) caches, and synthesizes new policies by retrieving relevant functions and grafting their KV caches when a new task is provided. Given retrieved function caches, FCGraft performs cache grafting via stitching, which composes cached function segments into a composite policy, and patching, which locally adapts only the necessary code regions to satisfy task-specific parameters and constraints with minimal additional decoding. By eliminating redundant prefill computation, this approach reduces generation latency, while reusing validated control structures improves robustness over prompt-level caching methods RAGCache, achieving 18.31% higher task success rate and 2.3x faster policy synthesis.

阅读与讨论 → 访问原文 →

02.

arXiv (CS.CL) 2026-06-16 DOI: arXiv:2605.21850

ACC: Compiling Agent Trajectories for Long-Context Training

作者:

Qisheng Su↗Zhen Fang↗Shiting Huang↗Yu Zeng↗Yiming Zhao↗Kou Shi↗Ziao Zhang↗Lin Chen↗Zehui Chen↗Lijun Wu↗Feng Zhao↗

Recent development of agents has renewed demand for long-context reasoning capacity of LLMs. However, training LLMs for this capacity requires costly long-document curation or heuristic context synthesis. We observe that agents produce massive trajectories when solving problems, invoking tools and receiving environment observations across many turns. The evidence needed to answer the original question is thus scattered throughout these turns, requiring integration of distant context segments. Nevertheless, standard agent SFT masks tool responses and only trains turn-level tool selection, creating a supervision blind spot where these scattered signals go unused. We propose Agent Context Compilation (ACC), which converts trajectories from search, software engineering, and database querying agents into long-context QA pairs that combine the original question with tool responses and environment observations gathered across multiple turns, training the model to answer directly without tool use. This makes the dependencies between the question and the evidence explicit, enabling direct supervision of long-context reasoning over distant segments without additional annotation. ACC is a simple but effective approach that can be combined with any existing long-context extension or training method, providing scalable supervised fine-tuning data. We validate ACC on long-range dependency modeling tasks through MRCR and GraphWalks, challenging benchmarks requiring cross-turn coreference resolution and graph traversal over extended contexts. Training Qwen3-30B-A3B with ACC achieves 68.3 on MRCR (+18.1) and 77.5 on GraphWalks (+7.6), results comparable to Qwen3-235B-A22B, while preserving general capabilities on GPQA, MMLU-Pro, AIME, and IFEval. Further mechanism analysis reveals that the ACC-trained model exhibits task-adaptive attention restructuring and expert specialization.

阅读与讨论 → 访问原文 →

03.

arXiv (CS.LG) 2026-06-17 DOI: arXiv:2606.18066

NoiseTilt: Noise-Tilted Reverse Kernels for Diffusion Reward Alignment

作者:

Jisung Hwang↗Yunhong Min↗Jaihoon Kim↗I-Chao Shen↗Minhyuk Sung↗

arXiv:2606.18066v1 Announce Type: new Abstract: We introduce the Noise-Tilted Reverse Kernel (NTRK), a reward-guided diffusion sampler that injects reward gradients through the noise term, leaving the pretrained reverse kernel unchanged and requiring only a single sample per step. Reward-guided sampling at inference time has greatly expanded the versatility of pretrained diffusion models. Yet existing methods face a trade-off. Gradient-based guidance shifts the reverse mean, steering generation but pushing intermediate states outside the region that the model was trained on and degrading quality. Search-based methods preserve quality but gain no gradient signal. No prior method achieves both. NTRK resolves this by keeping the reverse mean fixed and biasing the noise term toward high reward. We introduce a whitening operator, the central mechanism behind NTRK, that makes the reward gradient safe to inject as noise without losing its guiding signal. Across various reward alignment tasks, NTRK outperforms recent state-of-the-art baselines without losing sample quality. Remarkably, on aesthetic generation, NTRK surpasses the reward of the best baseline at 500 NFEs using only 25 NFEs, a 20$\times$ reduction in compute.

阅读与讨论 → 访问原文 →

04.

arXiv (CS.CL) 2026-06-11 DOI: arXiv:2606.11599

When is Your LLM Steerable?

作者:

Chenrui Fan↗Yize Cheng↗Ming Li↗Soheil Feizi↗Tianyi Zhou↗

Activation steering offers a lightweight approach to control language models' behavior at inference time, but whether it succeeds or fails heavily depends on the prompt, concept, model, and steering configuration. Finding the regime and boundaries of successful steering typically requires expensive grid searches and post-hoc evaluation of full autoregressive rollouts. In this work, we investigate whether steerability can be predicted from the model's internal states at the beginning of the generation process, e.g., after generating the first few tokens, and how to leverage such a predictor to improve steering success rate. To this end, we first introduce ASTEER, a testbed including 1.4M steered generations, spanning 150 concepts with each steering success/failure labeled. Leveraging this testbed, we analyze the model's early decoding dynamics by extracting features that compare hidden states before and after steering across layers and initial decoding steps. These features help us understand how steering's effects propagate along layers and token positions, which provide key information for steerability prediction. We then train a Gradient Boosting Decision Trees (GBDT) classifier on these features to predict whether an intervention will under-steer, succeed, or over-steer without requiring full rollout. Our predictor achieves around 0.7 macro-F1 score on unseen concepts, demonstrating that early hidden states encode substantial, structured information about eventual steering efficacy. We further leverage this steerability predictor as guidance for steering strength searching, achieving near-optimal performance with a small fraction of decoding cost.

阅读与讨论 → 访问原文 →

05.

arXiv (CS.CL) 2026-06-24 DOI: arXiv:2606.24667

DREAM: Dense Retrieval Embeddings via Autoregressive Modeling

作者:

Yixuan Tang↗Yi Yang↗

Dense retrieval embedding models are a fundamental component of modern retrieval-based AI systems. Most dense retrievers are trained with contrastive objectives, which require labeled positive and negative document pairs that are often costly and difficult to obtain. In this work, we investigate whether the autoregressive next-token prediction objective of a large language model (LLM) can provide supervision for dense retrieval. The intuition is simple: if a document contains information relevant to a query, conditioning on that document should make the target output easier for the LLM to predict. A key challenge is that the next-token prediction loss is computed inside the LLM, while the retriever is a separate embedding model. To address this challenge, we propose DREAM (Dense Retrieval Embeddings via Autoregressive Modeling), which injects retriever-generated query-document similarity scores into selected attention heads of a frozen LLM. During training, these scores determine how much attention each candidate document receives while the LLM predicts the target output. The resulting prediction loss provides gradients for retriever training through the attention mechanism. We evaluate DREAM on retrieval benchmarks BEIR and RTEB using embedding backbones ranging from 0.5B to 3B parameters. DREAM consistently outperforms existing baselines across different model scales. These results demonstrate that DREAM provides a promising approach for training dense retrievers through autoregressive modeling.

阅读与讨论 → 访问原文 →

06.

bioRxiv (Bioinfo) 2026-06-20 DOI: HASH:5f66831cf7e2196787d9c4554883b964

RNAStabFormer: Region-Aware Multi-Task Hybrid Learning for RNA Stability Prediction from Pulse-Chase Transcriptomics

作者:

Wang↗Zhang↗

RNA stability is a central layer of post-transcriptional gene regulation, yet large-scale stability labels derived from pulse-chase transcriptomics depend strongly on quantification region, time-window definition, and replicate quality control. We present RNAStabFormer, a controlled learning framework for predicting human RNA stability proxies from transcript sequence. Its core model, RAMHT, combines region-specific nucleotide Transformer encoders for CDS, and sequence, a CDS codon stream, engineered sequence-grammar features, gated fusion, and four task-specific regression heads. We construct four strict consensus labels from ENCODE BrU-seq/BruChase-seq data by crossing gene-sense and exon-sense quantification with late-chase 6 h/2 h and total-chase 6 h/0 h retention ratios, and evaluate all models on fixed repeated-random and chromosome-holdout splits. Across chromosome holdouts, XGBoost remains the strongest standalone model, with median Pearson correlations of 0.504, 0.544, 0.546, and 0.778 on the four labels. RAMHT is competitive with raw-sequence deep models but does not universally exceed engineered-feature baselines. A strict nested RAMHT–XGBoost blend nevertheless improves gene total-chase prediction by 0.017 mean Pearson and exon late-chase prediction by 0.004 mean Pearson over XGBoost. Region and mechanism analyses show that CDS, local k-mer composition, and codon-sensitive signals dominate predictive information. RNAStabFormer therefore provides both a multi-task neural model and a leakage-controlled evaluation protocol for RNA stability prediction from pulse-chase data.

阅读与讨论 → 访问原文 →

07.

arXiv (CS.AI) 2026-06-24 DOI: arXiv:2606.20737

Repeated Shared Access Enables Grokking, but Edit Propagation Depends on an Addressable Memory

作者:

Yanan Niu↗

arXiv:2606.20737v2 Announce Type: replace Abstract: We study factual edit propagation in a controlled synthetic knowledge-graph QA setting using a 2x2 grid that crosses loop recurrence with shared-memory access: a dense transformer (Dense), a looped transformer (Loop), a dense backbone with shared memory (Dense+Mem), and a looped backbone with shared memory (loop-memory coupling, LMC). The two factors dissociate. For learning, both routes to repeated shared access – looped recomputation and repeated memory rereading – cross the out-of-distribution (OOD) grokking barrier that Dense fails, so repeated shared access is the behavioral regularity, not a specific architecture. For editing, the substrates split along a different axis: applying a single localized factual edit (conditioned on direct success) and measuring 2-hop propagation on a shared pre-edit-correct set, the edit propagates strongly in both memory-bearing cells (LMC 0.78-0.92, Dense+Mem 0.71-0.96) and only weakly in the memory-free ones (Loop 0.04-0.30, Dense 0.00-0.03). The split is along the memory axis, not the loop axis: every memory-bearing seed exceeds every memory-free seed, with no detectable difference between the two memory cells. Crucially Dense+Mem has no recurrence, so the propagating ingredient is an addressable site that an edit can write to and later computation rereads, not loop recomputation; Loop is at best a partial intermediate. The affordance survives coarsening the store (N=128 to N=13): propagation attenuates but the memory/no-memory split persists, so fine granularity buys precision rather than the affordance itself. These results dissociate learning competence from editing affordance – repeated shared access suffices to grok, but edit propagation depends on whether the substrate exposes an addressable memory that the forward computation can write to and later reread, an affordance that loop recurrence provides only partially.

阅读与讨论 → 访问原文 →

08.

arXiv (quant-ph) 2026-06-16 DOI: arXiv:2606.15935

Experimental Observation of Dynamical Phase Transitions in a Dephased Photonic Quantum Walk

作者:

Xiaojian Huang↗Lei Xiao↗Bingzi Huo↗Xiaowei Wang↗Stefano Longhi↗Peng Xue↗

arXiv:2606.15935v1 Announce Type: new Abstract: Dynamical phase transitions in open quantum systems govern how non-equilibrium states relax toward a stationary state. We study these transitions experimentally using a discrete-time photonic quantum walk on a three-node graph. A tunable synthetic gauge flux and calibrated dephasing allow us to control time-reversal symmetry and the detailed balance properties of the effective Markovian dynamics. With detailed balance, we observe a first-order dynamical phase transition marked by a crossing of real Liouvillian eigenvalues. When detailed balance is broken, we observe a second-order dynamical phase transition at an exceptional point where eigenvalues and eigenvectors coalesce. By progressively reducing the dephasing strength, we track the crossover toward the quantum-coherent regime and determine that the transitions persist down to a finite threshold. Our results link Liouvillian spectral topology to relaxation criticality and demonstrate a controllable platform for engineered dissipative dynamics.

阅读与讨论 → 访问原文 →

09.

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 →

10.

arXiv (CS.CV) 2026-06-15 DOI: arXiv:2606.13896

How do Self-Supervised Remote Sensing Vision Models Transfer to Downstream Tasks?

作者:

Julia Romero↗Qin Lv↗Morteza Karimzadeh↗

Self-supervised geospatial foundation models (GeoFMs) learn transferable representations from remote sensing data, but their downstream behavior is difficult to characterize. We study six representative GeoFMs spanning joint-embedding, reconstruction, and multimodal pretraining families, and evaluate transfer across classification, regression, and segmentation benchmarks under different label availability and downstream pipelines. We find that model rankings change across tasks and adaptation settings. Layerwise probing shows that, in most cases, task-relevant information is more accessible in intermediate transformer blocks compared to final-layer embeddings, and that GeoFMs exhibit distinct depthwise profiles. In segmentation case studies on PASTIS and Sen1Floods11, downstream adaptation settings such as decoder design and fine-tuning can be as impactful as the choice of GeoFM, and standard dense-prediction heads may be poorly aligned with how GeoFMs organize information over depth. Finally, CKA analysis on case studies shows that fine-tuning does not rewrite GeoFMs uniformly across depth, and the strongest changes are localized to the first linear layer of the MLP in ViT blocks. These results help explain why GeoFM rankings shift across benchmarks and motivate more representation-aware evaluation and adaptation strategies.

阅读与讨论 → 访问原文 →

11.

PLOS Medicine 2026-06-23 DOI: HASH:69d897c60dd4043f444525a07759c603

Parental body mass index and offspring childhood body size and eating behaviour: A structural equation modelling analysis in the Norwegian Mother, Father and Child Cohort Study

作者:

Tom A. Bond↗

by Tom A. Bond, Tom A. McAdams, Nicole M. Warrington, Laurie J. Hannigan, Espen Moen Eilertsen, Ziada Ayorech, Fartein A. Torvik, George Davey Smith, Deborah A. Lawlor, Eivind Ystrom, Alexandra Havdahl, David M. Evans Background The intergenerational transmission of obesity-related traits could propagate an accelerating cycle of obesity, if parental adiposity causally influences offspring adiposity. The extent to which intergenerational obesity associations are due to such causal effects, as opposed to genetic confounding (inheritance), is unclear. We aimed to establish whether associations between parental peri-pregnancy body mass index (BMI) and offspring birth weight (BW), BMI until 8 years of age, and 8-year-old eating behaviour are due to genetic confounding. Methods and findings Data were from the Norwegian Mother, Father and Child Cohort Study, a prospective population-based birth cohort born between 1999 and 2009 at 50 out of 52 hospital maternity units in Norway. We compared the strength of the associations of maternal pre-pregnancy BMI versus paternal BMI during pregnancy, with offspring outcomes including birth weight and BMI assessed between age 6 months and 8 years of age, and appetite-related eating behaviour traits assessed at age 8 years via the Child Eating Behaviour Questionnaire (CEBQ), adjusting for potential confounders including parity, parental/grandparental language group and parental age, smoking, education and income). We then used an extended children of twins structural equation model (SEM) to quantify the extent to which associations were due to genetic confounding. Up to 85,866 children (51.3% male) were included in linear regression models, whereas SEM models included up to 50,999 children. Maternal BMI was more strongly associated than paternal BMI with offspring BW, but the maternal-paternal difference decreased for offspring BMI after birth. Greater parental BMI was associated with obesity-related offspring eating behaviours. SEM results indicated that genetic confounding did not explain the association between parental BMI and offspring BW, but explained the majority of the association with offspring BMI from 6 months onwards. For 8-year BMI, genetic confounding explained 79% (95% CI [62, 95]; p = 1.9 × 10−12) of the covariance with maternal BMI and 94% (95% CI [72, 113]; p = 2.7 × 10−14) of the covariance with paternal BMI. Limitations of this study include selective recruitment and attrition, potential bias due to parental assortative mating, and that findings may not generalise beyond high-income country settings with high obesity prevalence. Conclusions We found strong evidence that parent–child BMI associations may primarily be due to genetic confounding. When considered alongside prior evidence, this finding may argue against a strong causal effect of maternal or paternal adiposity on childhood adiposity via intrauterine or periconceptional mechanisms.

阅读与讨论 → 访问原文 →

12.

arXiv (CS.AI) 2026-06-12 DOI: arXiv:2606.13176

Mental-R1: Aligning LLM Reasoning for Mental Health Assessment

作者:

Xin Wang↗Boyan Gao↗Yibo Yang↗David A. Clifton↗

arXiv:2606.13176v1 Announce Type: new Abstract: Mental health problems such as anxiety, depression, and suicide remain urgent global challenges, where timely and accurate assessment is critical for effective intervention. Recently, large language models have been explored for mental health assessment. However, existing general-purpose post-training methods do not align with the cognitive processes of human assessment, which may lead to unreliable reasoning outcomes. To bridge this gap, we propose Cognitive Relative Policy Optimization (CRPO), a reinforcement learning framework tailored for the mental health domain. CRPO extends group relative policy optimization by integrating stage-dependent uncertainty modeling into the policy optimization process. Specifically, we introduce a stage-wise entropy regularization mechanism that encourages broad exploration in early reasoning phases and progressively enforces confident decision-making in later stages, mimicking the human cognitive shift from uncertainty to certainty. In addition, inspired by cognitive appraisal theory, we formalize cognitive reasoning stages, thereby guiding theory-grounded interpretable inference. Experiments on 8 mental health datasets show that CRPO achieves an average improvement of 10.4 percentage points in weighted F1-score over the best reinforcement learning baseline. Furthermore, the CRPO-trained model Mental-R1 demonstrates clear advantages compared with existing large language models on reasoning-intensive cases, suggesting that CRPO enhances reasoning capabilities for mental health assessment.

阅读与讨论 → 访问原文 →

13.

arXiv (CS.AI) 2026-06-18 DOI: arXiv:2606.18688

Dual-Channel Grounded World Modeling (DCGWM): Structural Prevention of Objective Interference Collapse via Heterogeneous External Grounding with Inward-Only Gradient Flow

作者:

Akshay Hazare↗

arXiv:2606.18688v1 Announce Type: cross Abstract: Joint Embedding Predictive Architectures (JEPAs) are a leading approach to world model representation learning. We identify a failure mode in JEPA-based world models grounded against two qualitatively distinct external signals: physical dynamics (sparse, high-magnitude, constraint-satisfying gradient corrections) and social-behavioral dynamics (diffuse, distribution-matching corrections). We term this Objective Interference Collapse (OIC): we argue that joint learning in a shared latent space causes the dominant channel to systematically collapse the subordinate channel's representational subspace, in a manner not resolvable by loss weighting alone. We propose Dual-Channel Grounded World Modeling (DCGWM), designed to structurally prevent OIC through a partitioned latent space (physical subspace Z_p, behavioral subspace Z_b) with inward-only gradient flow. A Physical Grounding Channel updates only Z_p via VICReg-style alignment to physical measurements; a Social-Behavioral Grounding Channel updates only Z_b via alignment to trajectories from an emergent multi-agent simulation. An Inter-Channel Interface Module couples the subspaces at the task level without cross-subspace gradients. An Asymmetric Grounding Adherence Loss penalizes rollout drift with a hard hinge for physical violations and a soft KL for behavioral divergence. A Generative Rendering Layer is architecturally isolated from the latent world model. We present three theoretical results: the partition removes the gradient-interference pathway implicated in OIC; each grounded subspace inherits anti-collapse guarantees from its alignment objective; and generative isolation is necessary under a stated assumption on the generative objective's geometry. This manuscript establishes the problem formulation and architecture; experimental validation is ongoing and will be reported in a future revision.

阅读与讨论 → 访问原文 →

14.

arXiv (CS.AI) 2026-06-15 DOI: arXiv:2602.05670

HyperPotter: Spell the Charm of High-Order Interactions in Audio Deepfake Detection

作者:

Qing Wen↗Haohao Li↗Zhongjie Ba↗Peng Cheng↗Miao He↗Li Lu↗Kui Ren↗

arXiv:2602.05670v2 Announce Type: replace-cross Abstract: Advances in AIGC technologies have enabled the synthesis of highly realistic audio deepfakes capable of deceiving human auditory perception. Although numerous audio deepfake detection (ADD) methods have been developed, most rely on local temporal/spectral features or pairwise relations, overlooking high-order interactions (HOIs). HOIs capture discriminative patterns that emerge from multiple feature components beyond their individual contributions. We propose HyperPotter, a hypergraph-based framework designed to capture high-order relations associated with synergistic patterns through clustering-based hyperedges with class-aware prototype initialization. Extensive experiments on 13 test sets show that HyperPotter improves over the baseline on 11 sets, yielding an average relative EER reduction of 12.68\% across all test sets and 22.15\% on the improved sets. These results demonstrate strong cross-scenario generalization, while also revealing robustness limits under severe codec or channel distortion.

阅读与讨论 → 访问原文 →

15.

arXiv (CS.LG) 2026-06-15 DOI: arXiv:2604.04611

Dynamic Free-Rider Detection in Federated Learning via Simulated Attack Patterns

作者:

Motoki Nakamura↗

arXiv:2604.04611v2 Announce Type: replace Abstract: Federated learning (FL) enables multiple clients to collaboratively train a global model by aggregating local updates without sharing private data. However, FL often faces the challenge of free-riders, clients who submit fake model parameters without performing actual training to obtain the global model without contributing. Chen et al. proposed a free-rider detection method based on the weight evolving frequency (WEF) of model parameters. This detection approach is a leading candidate for practical free-rider detection methods, as it requires neither a proxy dataset nor pre-training. Nevertheless, it struggles to detect ``dynamic'' free-riders who behave honestly in early rounds and later switch to free-riding, particularly under global-model-mimicking attacks such as the delta weight attack and our newly proposed adaptive WEF-camouflage attack. In this paper, we propose a novel detection method S2-WEF that simulates the WEF patterns of potential global-model-based attacks on the server side using previously broadcasted global models, and identifies clients whose submitted WEF patterns resemble the simulated ones. To handle a variety of free-rider attack strategies, S2-WEF further combines this simulation-based similarity score with a deviation score computed from mutual comparisons among submitted WEFs, and separates benign and free-rider clients by two-dimensional clustering and per-score classification. This method enables dynamic detection of clients that transition into free-riders during training without proxy datasets or pre-training. We conduct extensive experiments across three datasets and five attack types, demonstrating that S2-WEF achieves higher robustness than existing approaches.

阅读与讨论 → 访问原文 →

16.

arXiv (CS.LG) 2026-06-15 DOI: arXiv:2606.14029

Utility-Constrained Policy Optimization

作者:

Mehrdad Moghimi↗Bernardo Avila Pires↗

arXiv:2606.14029v1 Announce Type: new Abstract: Constrained MDPs (CMDPs) are a widely adopted framework for incorporating safety into RL agents; however, the framework does not support risk-sensitive constraints. This can be problematic: For example, CMDPs allow for optimal solutions that, in order to satisfy the risk-neutral constraints, mix infrequent catastrophic behaviors and frequent, overly conservative ones. Moreover, prior empirical results suggest that enforcing stricter, risk-sensitive constraints can improve performance even under risk-neutral evaluation. The natural framework to incorporate risk-sensitive constraints is utility-constrained MDPs (UCMDPs), but no practical solutions for this problem existed. In this work, we introduce a simple yet powerful methodology for UCMDPs and constrained RL. Besides allowing for risk-sensitive constraints, our framework does not require us to fix constraint limits in advance of training the agent, provided that a sensible range is known. This increases policy flexibility and, in practice, allows for adjustments to these limits at no extra training cost. Besides benefiting from the generality of the framework, our agent shows strong performance in practice, consistently matching or outperforming existing baselines in several Safety Gymnasium benchmark tasks.

阅读与讨论 → 访问原文 →

17.

arXiv (quant-ph) 2026-06-19 DOI: arXiv:2606.20281

Arrival times of an atomic Bose-Einstein condensate

作者:

Pascal Naidon↗Lucas Happ↗Denis Boiron↗

arXiv:2606.20281v1 Announce Type: cross Abstract: The times of flight of an atomic Bose-Einstein condensate are theoretically investigated in the experimentally unexplored regime corresponding to detection close to the trap of the condensate. In this regime, there is no consensus on how to calculate the distribution of times of arrival onto the detector. For non-interacting particles, distinct theoretical predictions have been made in the past. This work analyses how these predictions are modified for an interacting Bose-Einstein condensate. For this purpose, a time-dependent Gross-Pitaevskii equation is solved analytically and numerically.

阅读与讨论 → 访问原文 →

18.

arXiv (CS.AI) 2026-06-19 DOI: arXiv:2606.19721

OnDeFog: Online Decision Transformer under Frame Dropping

作者:

Daiki Yotsufuji↗Kenta Nishihara↗Shoma Shimizu↗Kento Uchida↗Shinichi Shirakawa↗

arXiv:2606.19721v1 Announce Type: cross Abstract: In challenging real-world reinforcement learning applications, communication delays or sensor failures often cause frame dropping, in which the agent cannot receive the dropped states and associated rewards. To address the performance degradation caused by frame dropping, the Decision Transformer under Random Frame Dropping (DeFog) was developed by incorporating additional mechanisms into the decision transformer to tackle frame dropping. Although DeFog can mitigate performance degradation in frame-dropping environments, since DeFog is an offline learning method, it struggles to effectively generalize to novel states not adequately represented in the training dataset. In this study, we propose OnDeFog, which integrates the mechanisms in DeFog with the online decision transformer (ODT), an online reinforcement learning method that learns policies through direct environmental interaction. Comprehensive experimental evaluation demonstrates that our proposed OnDeFog achieves superior performance compared to ODT in environments characterized by high dropping frame rate and outperforms DeFog on datasets containing a large amount of low-reward data.

阅读与讨论 → 访问原文 →

19.

arXiv (CS.LG) 2026-06-15 DOI: arXiv:2505.16319

FreshRetailNet-LT: A Stockout-Annotated Censored Demand Dataset for Latent Demand Recovery and Forecasting in Fresh Retail

作者:

Yangyang Wang↗Jiawei Gu↗Li Long↗Xin Li↗Li Shen↗Zhouyu Fu↗Xiangjun Zhou↗Xu Jiang↗

arXiv:2505.16319v4 Announce Type: replace Abstract: Accurate demand estimation is critical for the retail business in guiding the inventory and pricing policies of perishable products. However, it faces fundamental challenges from censored sales data during stockouts, where unobserved demand creates systemic policy biases. Existing datasets lack the temporal resolution and annotations needed to address this censoring effect. To fill this gap, we present FreshRetailNet-50K, the first large-scale benchmark for censored demand estimation. It comprises 50,000 store-product time series of detailed hourly sales data from 898 stores in 18 major cities, encompassing 863 perishable SKUs meticulously annotated for stockout events. The hourly stock status records unique to this dataset, combined with rich contextual covariates, including promotional discounts, precipitation, and temporal features, enable innovative research beyond existing solutions. We demonstrate one such use case of two-stage demand modeling: first, we reconstruct the latent demand during stockouts using precise hourly annotations. We then leverage the recovered demand to train robust demand forecasting models in the second stage. Experimental results show that this approach achieves a 2.73% improvement in prediction accuracy while reducing the systematic demand underestimation from 7.37% to near-zero bias. With unprecedented temporal granularity and comprehensive real-world information, FreshRetailNet-50K opens new research directions in demand imputation, perishable inventory optimization, and causal retail analytics. The unique annotation quality and scale of the dataset address long-standing limitations in retail AI, providing immediate solutions and a platform for future methodological innovation. The data (https://huggingface.co/datasets/Dingdong-Inc/FreshRetailNet-50K) and code (https://github.com/Dingdong-Inc/frn-50k-baseline}) are openly released.

阅读与讨论 → 访问原文 →

20.

arXiv (CS.CV) 2026-06-25 DOI: arXiv:2606.25491

HG-Bench: A Benchmark for Multi-Page Handwritten Answer-Region Grounding in Automated Homework Assessment

作者:

Chuangxin Zhao↗Boyan Shi↗Yanling Wang↗Yijian LU↗Canran Xiao↗Jiali Chen↗Jun Xia↗Yan Wang↗Ji Qi↗Juanzi Li↗

Automated homework assessment depends not only on recognizing student answers, but also on accurately locating where each answer and each intermediate reasoning step appears in noisy, multi-page handwritten work. This paper addresses the missing evaluation setting of page-aware, two-level answer-region grounding: given a sequence of homework page images, a model must localize complete answer regions and their ordered step-level subregions. We introduce HG-Bench, a benchmark of 500 human-annotated K-12 homework samples curated from a 1,489,278-image source pool, with question-level and step-level boxes linked by a hierarchical containment constraint. HG-Bench is paired with a page-aware evaluation protocol that separately measures complete-answer localization (FA) and step-level decomposition (FSm), revealing whether models truly ground the spatial structure of student reasoning rather than merely parse visible text. Across frontier closed-source APIs and competitive open-weight VLMs, no zero-shot system exceeds 55.22% on FA or 48.22% on FSm, while a GLM-4.6V 9B reference model fine-tuned on ~10k in-domain examples reaches 74.97/72.26. These results identify step-level handwritten grounding as a concrete capability gap and provide a reproducible benchmark, evaluation protocol, and trained reference point for future work on automated homework assessment.

阅读与讨论 → 访问原文 →

21.

arXiv (CS.AI) 2026-06-16 DOI: arXiv:2606.14790

XFlow: An Executable Protocol Programming System for Reliable Multi-Agent Workflows

作者:

Hanqi Li↗Jing Peng↗Zijian Wang↗Lu Chen↗Kai Yu↗

arXiv:2606.14790v1 Announce Type: cross Abstract: LLM-based multi-agent systems increasingly coordinate planning, reasoning, tool use, and human interaction, yet their reliability remains limited. A central source of this limitation is the underspecified prompt–harness boundary. Current systems lack a principled way to decide which workflow commitments should remain in prompts and which should become harness structure. We present XFlow, an executable protocol programming system for reliable multi-agent workflows, and XPF (XFlow Protocol Format), its domain-specific protocol programming language. XFlow occupies a middle position between prompt-only orchestration and markup-like workflow descriptions. XPF remains readable as a literate protocol, but it is compiled and executed as a program. Its design keeps informal semantic work inside actors while moving selected commitments into harness structure that can be checked, preserved, and enforced. At runtime, XFlow stages uncertainty through lifecycle-governed symbols, which are typed state cells with validation and commit states. Actor outputs are mediated before they become shared state, instead of spreading through prompts, transcripts, or implicit memory. Our experiments cover Constrained Interaction, Long-Context Reasoning, and Agentic Software Engineering. They show that XFlow improves reliability by making constraints, evidence handling, and process requirements explicit and enforceable.

阅读与讨论 → 访问原文 →

22.

arXiv (CS.AI) 2026-06-15 DOI: arXiv:2606.14570

Regional Climate Model Emulation with Diffusion Approaches: What is the Added Value of Generative Machine Learning?

作者:

Mikel N. Legasa↗Antoine Doury↗Achille Gellens↗Redouane Lguensat↗Clara Naldesi↗Soulivanh Thao↗Mathieu Vrac↗

arXiv:2606.14570v1 Announce Type: cross Abstract: Emulators provide a cost-effective alternative to regional climate models (RCMs) by capturing their dynamical downscaling function. They link large-scale predictors simulated by global climate models (GCMs) to RCM-simulated high-resolution fields of the target variable, here precipitation. Machine learning methods, typically deep learning, are cheaper than running RCMs in computation time and energy. Among them, generative models are appealing because they can simulate ensembles of local high-resolution fields consistent with the predictors. This ensemble, which we call the uncertainty envelope, remains to be properly assessed for added value. Here, we make three contributions. First, we introduce ParamDiffusion, a new two-stage diffusion-based framework, and compare it with a state-of-the-art diffusion approach. Second, we expand standard validation through a comprehensive framework aligned with climate-science needs, examining specific precipitation events, including extremes. Third, within this framework, we assess the added value of diffusion approaches relative to deterministic methods. We intercompare four deep-learning models: a deterministic model designed to capture the precipitation tail; a parametric probabilistic model based on it; a recently proposed diffusion approach; and ParamDiffusion, which couples the parametric model with a diffusion model. Our results show that diffusion-based approaches reproduce climatological precipitation statistics with high skill, including distributional tails and spatially compounded extremes, while generating spatially detailed fields. However, none of the assessed models consistently accounts for the most extreme RCM-simulated events within its uncertainty envelope. Diffusion models are therefore promising for probabilistic RCM emulation, but progress is still required before they can reliably represent high-impact precipitation extremes.

阅读与讨论 → 访问原文 →

23.

bioRxiv (Bioinfo) 2026-06-11 DOI: HASH:e07849d17e38c6df1247ffe86e29d25d

Hyper3D-lite: count-preserving representation auditing for long-read multi-contact genome data

作者:

Zhang↗

Long-read and single-molecule sequencing technologies are rapidly increasing molecule-level data, with platforms such as Oxford Nanopore, PacBio HiFi, and Roche sequencing-by-expansion advancing at different technology readiness levels. In the specific context of Pore-C and HiPore-C multi-contact chromatin-conformation assays, long-read multi-contact 3D genome assays preserve molecule-level contact context, but common downstream pairwise projections can expand one multi-contact molecule into many pair records. This creates a representation problem: apparent contact evidence can increase through the counting frame before biological interpretation begins. Hyper3D-lite addresses this problem as a representation-first audit tool for read-to-fragment-style long-read multi-contact inputs. It compares all-pair projection with CPB, a count-preserving statistical accounting reference point, and separates broad software outputs from conservative higher-order candidate calls.

阅读与讨论 → 访问原文 →

24.

bioRxiv (Bioinfo) 2026-06-23 DOI: HASH:da659647f04238e31eecb2a2c1476345

biomeStat: Using Agentic AI for Scalable Genomic Epidemiology Demonstrated Through End-to-End Analysis of 1,000 Asian Dengue Virus Genomes

作者:

Ariyaratne↗Somaratna↗Malavige↗G. N↗

Genomic epidemiology workflows typically require expert curation of multiple specialized tools, extensive manual parameter tuning, and access to heterogeneous compute infrastructure. While standard generative AI models often hallucinate in complex biological domains, we introduce biomeStat: an autonomous AI agent that functions as a strict deterministic orchestrator. By automatically writing code to execute established bioinformatics tools in sandboxed environments, biomeStat dynamically provisions compute resources (CPU and GPU) and guarantees reproducibility, making it immediately useful for scientists without requiring command-line expertise. To demonstrate the platform, we performed a fully autonomous genomic epidemiology and structural analysis of 1,000 Dengue virus (DENV) genomes sampled from 16 Asian countries between 2000 and 2025. The agent seamlessly orchestrated phylogenetic reconstruction (IQ-TREE, TreeTime), Bayesian phylodynamics (BEAST2 via NVIDIA H200 GPU), selection pressure analysis (HyPhy), and structural mapping (PyMOL). The analysis was completed in under 24 hours of wall-clock time, revealing endemic stability (R_e ~1.0) and identifying 1,869 candidate immune escape sites structurally colocalized with B-cell and T-cell epitopes. Furthermore, the agent validated 176 highly conserved drug target residues across the viral replication complex, confirming that resistance-associated positions for emerging antivirals JNJ-1802 and NITD-688 remain absolutely conserved across all four serotypes. By bridging the gap between natural language intent and deterministic computational execution, biomeStat reduces weeks of expert effort into a single-session analysis with full methodological transparency.

阅读与讨论 → 访问原文 →

25.

medRxiv (Medicine) 2026-06-23 DOI: HASH:3e12669e992120b285bda34c261656ed

Shared Polygenic Architecture Across Arteriopathies: An Integrative Cross-Trait Analysis

作者:

Brennan↗S. O↗CADISP Consortium↗Tinworth↗A. C↗Daghlas↗Le Grand↗Rioux↗Kelly↗P. J↗Gill↗Debette↗…

Background: Non-monogenic arteriopathies are often classified as distinct entities according to the arterial territory involved, yet they share clinical features and may co-occur in the same individual. This pattern suggests shared susceptibility across anatomically distinct arteriopathies, potentially driven by common biological and genetic mechanisms. Methods: We investigated the shared genetic architecture of five arteriopathies (cervical artery dissection (CeAD), intracranial aneurysm (IA), spontaneous coronary artery dissection (SCAD), aortic aneurysm and dissection (AAD), and fibromuscular dysplasia (FMD)) using LD score regression, Association analysis based on SubSETs (ASSET), pairwise Multi-Trait Analysis of Genome-wide association summary statistics (MTAG), pleiotropy mapping and Mendelian randomization (MR) to identify shared loci and prioritise candidate causal genes. Results: LD score regression identified significant positive genetic correlations between CeAD-SCAD (rg = 0.64), IA-AAD (rg = 0.33), IA-SCAD (rg = 0.37), CeAD-AAD (rg = 0.56) and SCAD-AAD (rg = 0.20). ASSET identified 37 shared independent loci, and in MTAG analyses, one novel locus was identified for CeAD and SCAD (SLC39A8) and one for IA (FGF5). 13 loci showed strong cross-trait colocalization, including PHACTR1, LRP1, and CDKN2B-AS1. Using the Genotype-Phenotype Map, we found that arteriopathy-associated variants colocalized with blood pressure- and migraine-related traits, while many showed effect directions opposite to those observed for coronary artery disease. Proteome-wide MR identified 67 circulating proteins associated with at least one trait, including ECM1 and SHISA5 for CeAD and FGF5 for IA, with 17 supported by colocalization. Transcriptome-wide MR identified 204 colocalized tissue?specific signals, of which, 14 were shared across multiple traits. Enrichment analyses implicated pathways related to vascular development, smooth muscle cell function, extracellular matrix organization, and TGF-? signaling. Conclusions: These findings support shared genetic architecture across anatomically distinct arteriopathies, implicating pathways involved in vascular structure and prioritising therapeutic targets for future mechanistic investigation.

阅读与讨论 → 访问原文 →