Academic Intelligence · Curated Daily

探索全球前沿学术脉络

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

01.
medRxiv (Medicine) 2026-06-22

''Circumstantial Determinants'': An Efficient Approach to Reaching People in Need of HIV Prevention?

HIV prevention and testing programmes primarily reach people who self-refer or attend routine health services. Higher-risk individuals are missed if they are healthy, under-estimate their risk of infection or under-report sexual risk-behaviours. We assess a new approach to address limitations in existing programmes by targeting HIV services on ''Circumstantial Determinants'' (CDs) of HIV risk - the social circumstances, settings, and norms associated with behaviours that increase risk of HIV acquisition. Data on potential CDs and sexual behaviour were collected in a population survey in Zimbabwe in 2018/19 (N=9141). HIV-negative individuals reporting [≥] 1 sexual risk-behaviours were defined as the 'priority population' for HIV prevention. For each sex, six circumstantial determinants were associated with being in the priority population (aOR [≥] 1.30; p [≤] 0.01). Reach and efficiency of CDs (and combinations) were calculated; ROC curve algorithms evaluated their ability to identify priority population membership; and HIV prevention condom cascades were compared between CD-defined priority population subgroups. Example findings include that targeting men at bars and beerhalls could reach 48.5% of the priority population and 25.1% of lower-risk men. These percentages increase to 77.1% and 53.7% if men with poor mental health, no religious affiliation, negative social capital, or living on agricultural estates are also targeted. Targeting women with poor mental health could reach 32.0% of the priority population and 21.3% of lower-risk women. Targeting additional circumstantial determinants increases these percentages to 54.1% and 37.5%, respectively. Cascade barriers to condom use differed between CD-defined subgroups. The Circumstantial Determinants approach demonstrates proof-of-concept potential to strengthen HIV prevention services.

02.
arXiv (CS.AI) 2026-06-24

Agentic AI for Bilevel Long-Term Optimization of Policy-Driven Physical Layer Systems

arXiv:2606.24416v1 Announce Type: new Abstract: Network operators' changing policies, service requirements, and stringent real-time constraints render existing methods designed with fixed objectives and constraints ineffective. This paper presents Agentic long-term performance optimization (Agentic-LTPO), a nested bilevel optimization framework that can be applied to adaptive physical layer problem configuration. The key idea is to employ agentic AI to generate upper-level configurations in a bilevel optimization structure, where evolving operator policies, environment summaries, and historical experiences are translated into structured lower-level optimization problem configurations. The lower level solves the problems with updated configurations for real-time physical-layer decisions. Considering cell-free MIMO beamforming as a use case, we embody Agentic-LTPO by designing a new multi-agent decision process with retrieval-augmented experience-based verification in the upper level, together with a closed-form beamformer in the lower level. Experiments demonstrate that Agentic-LTPO exhibits strong adaptability to dynamic operator policies and effectively enhances the system's long-term performance by 57.2% compared to traditional methods.

04.
arXiv (CS.AI) 2026-06-12

Meta-Learning Transformers to Improve In-Context Generalization

arXiv:2507.05019v2 Announce Type: replace-cross Abstract: In-context learning enables transformer models to generalize to new tasks based solely on input prompts, without any need for weight updates. However, existing training paradigms typically rely on large, unstructured datasets that are costly to store, difficult to evaluate for quality and balance, and pose privacy and ethical concerns due to the inclusion of sensitive information. Motivated by these limitations and risks, we propose an alternative training strategy where we leverage a collection of multiple, small-scale, and domain-specific datasets. We empirically demonstrate that the increased quality and diversity of such data improve the generalization abilities of in-context learners beyond their training domain, while achieving comparable performance with models trained on a single large-scale dataset. We investigate this paradigm by leveraging meta-learning to train an in-context learner on the Meta-Album collection under several settings. Firstly, we show the performance in a controlled environment, where the test domain is completely excluded from the training knowledge. Secondly, we explore the robustness of these models to forgetting in a continual scenario where the information is accessible for a limited time. Finally, we explore the more challenging unsupervised scenario. Our findings demonstrate that transformers still generalize for in-context prediction when trained on a curated dataset collection while offering advantages in modularity and replaceability.

05.
arXiv (math.PR) 2026-06-24

Conditioning of incoherent sub-dictionaries sampled from a coherent dictionary

arXiv:2606.24323v1 Announce Type: new Abstract: Motivated by the desire to find a realistic and stable random model for $d$-dimensional signals, that are sparse in a transform-based and thus often coherent frame, such as a wavelet or a Gabor frame, we study the conditioning of incoherent sub-dictionaries sampled from a coherent dictionary, such as a unit norm frame. In particular, we show that if the sub-dictionary is selected via a coherence rejective Poisson sampling model, it is well-conditioned with high probability, as long as its expected size scales as $d/\log (K)$, where $K$ is the number of dictionary elements. The result is proved for the more general case of sampling quadratic sub-matrices from a real but not necessarily symmetric $K\times K$ matrix with zero diagonal, where coherence rejective sampling is defined via a symmetric mask, that acts as coherence substitute.

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

OSGuard: A Benchmark for Safety in Computer-Use Agents

arXiv:2606.15034v1 Announce Type: new Abstract: Computer-use agents are increasingly evaluated by whether they complete realistic desktop and web tasks. However, task success alone can miss failures in which an agent reaches the nominal goal through an unsafe shortcut. We introduce OSGuard, a dual-granularity benchmark suite for evaluating safety in computer-use agents under benign, unchanged user instructions. OSGuard contains an action-level benchmark for local guardrail decisions and a risk-augmented execution suite for end-to-end evaluation. The action-level benchmark consists of contextualized proposed actions labeled as allowed, unrelated, or unsafe, each judged relative to the original instruction and current interface state. The execution suite contains manually constructed OSWorld-derived task variants in which the original task remains achievable, but the environment is modified to introduce latent hazards such as destructive overwrites, etc. Each variant is paired with augmented evaluators that retain the original task-success criterion while adding explicit state-based safety invariants, allowing us to distinguish safe completions from unsafe completions that satisfy the nominal task objective. Our experimental results on OSGuard show that current multimodal guardrails can perform well on isolated action judgments, while risk-augmented execution exposes remaining gaps between local oversight and reliable end-to-end safety. This dual-granularity design enables more precise diagnosis of whether models can both recognize unsafe proposed actions and improve full-task safety when deployed as guardrails.

07.
arXiv (CS.AI) 2026-06-17

Transformer-Based Warm-Starting for Feasible and Optimal Terminal Approach to Tumbling Objects with Space Manipulators

arXiv:2606.17317v1 Announce Type: cross Abstract: Real-time trajectory generation for on-orbit robotic servicing is challenging due to the nonlinear coupling between spacecraft bus motion, manipulator dynamics, visibility cone, and trajectory-level safety constraints. This paper studies learning-based warm-starting for sequential convex programming (SCP) in the terminal approach of a space manipulator toward a tumbling target. The proposed framework decomposes the problem into a system center-of-mass translational planning stage and a coupled attitude–manipulator torque-allocation stage, and applies a causal transformer warm-start to the latter, which constitutes the dominant computational bottleneck. Linear and flow matching action decoders are compared under different action-chunking and training dataset sizes, and the resulting warm-starts are evaluated under both cost-optimal and feasibility projection using SCP. Across 300 held-out scenarios, the learned warm-start reduces the second-stage SCP iteration count by up to 28% and the runtime by 23% while preserving the final control-cost distribution. When the learned warm-starts are used for nonconvex feasibility projection, they nearly halve the runtime relative to cost-optimal SCP, while avoiding the catastrophic high-cost tail behavior observed when initialized heuristically. These results indicate that sequence-model warm-starts can improve both the computational efficiency and trajectory robustness of optimization-based terminal guidance for space manipulation.

08.
arXiv (quant-ph) 2026-06-16

Chiral Lattice Gauge Theories from Symmetry Disentanglers

arXiv:2601.04304v2 Announce Type: replace-cross Abstract: We propose a Hamiltonian framework for constructing chiral gauge theories on the lattice based on symmetry disentanglers: constant-depth circuits of local unitaries that transform not-on-site symmetries into on-site ones. When chiral symmetry can be realized not-on-site and such a disentangler exists, the symmetry can be implemented in a strictly local Hamiltonian and gauged by standard lattice methods. Using lattice rotor models, we realize this idea in 1+1 and 3+1 spacetime dimensions for $U(1)$ symmetries with mixed 't Hooft anomalies, and show that symmetry disentanglers can be constructed when anomalies cancel. As an example, we present an exactly solvable Hamiltonian lattice model of the (1+1)-dimensional "3450" chiral gauge theory, and we argue that a related construction applies to the $U(1)$ hypercharge symmetry of the Standard Model fermions in 3+1 dimensions. Our results open a new route toward fully local, nonperturbative formulations of chiral gauge theories.

09.
arXiv (CS.AI) 2026-06-16

IoT-Zoo: A Container-Based Framework for Heterogeneous IoT Device Profiles and Reproducible Traffic Capture

arXiv:2606.15653v1 Announce Type: cross Abstract: The validation of networking and security solutions for the Internet of Things (IoT) requires realistic and reproducible experimental data. However, existing platforms often achieve scalability by replicating a limited set of device types, which restricts profile diversity and fails to capture the heterogeneity of real-world IoT environments. In this paper, we present IoT-Zoo, a container-based testbed designed to support reproducible experimentation through heterogeneous, dataset-driven IoT device profiles. Built upon Containernet, IoT-Zoo automates the deployment of multi-domain scenarios and supports real application protocols such as MQTT and RTSP. The platform provides a single-command interface for environment provisioning and automated traffic capture (PCAP), enabling the generation of consistent traffic baselines and reducing the operational effort required to evaluate networking and security solutions.

10.
arXiv (math.PR) 2026-06-12

Sticky CIR process with potential: invariant measure and exact sampling

arXiv:2605.13648v4 Announce Type: replace Abstract: We study the sticky Cox–Ingersoll–Ross (CIR) process in one dimension, a diffusion on $[0,\infty)$ with a sticky boundary condition at the origin, arising as the marginal process in a sparse Bayesian inference framework based on Hadamard–Langevin dynamics. For the parameter range $\delta\in(1,2)$, in which the origin is accessible but not absorbing, we prove well-posedness of the process and uniqueness of its invariant measure, which is a mixture of a point mass at zero and a weighted gamma-type density on the interior. We derive an explicit Green's function for the resolvent in terms of confluent hypergeometric functions, and use this to construct an exact sampler for the invariant measure in the zero-potential case. For a non-trivial potential $G$, we establish existence and uniqueness of the tilted invariant measure via a Girsanov change of measure, and develop two sampling algorithms: a Metropolis–Hastings corrected sampler that targets the invariant measure exactly, and a cheaper, biased unadjusted Langevin algorithm (ULA) for a boundary-clamped variant of which we prove a first-order expansion of the stationary bias with an explicit constant: the leading error is a rank-one transfer of mass $K_\star h|\log h| $ onto the atom, so the total-variation bias is of exact order $h|\log h | $ – independent of $\delta$ – whenever the potential has nonzero boundary drift. Numerical experiments confirm the predicted behaviour: the Metropolis–Hastings sampler achieves the target invariant measure at all step sizes, while the ULA bias follows the proven first-order law, including its constant.

12.
arXiv (quant-ph) 2026-06-24

Resource theory of interactive quantum instruments

arXiv:2603.27676v2 Announce Type: replace Abstract: Quantum instruments describe both the classical outcome and the updated quantum state in a measurement process. To do this in a non-trivial way, instruments must have the capability to interact coherently with the state that they measure. Here, we develop a resource theory for instruments. We consider a relevant quantifier of the separation between interactive and non-interactive instruments and show that it admits three distinct operational interpretations in terms of quantum information tasks. These concern (i) the preservation of maximally entangled states after a local measurement, (ii) the average ability to preserve random states after measurement, and (iii) the ability to recover the classical information generated from measuring half of a maximally entangled state. We also introduce a natural set of allowed operations and show that the third task fully characterises the resource content of instruments. Our general framework reproduces as special cases established resource theories for channels and measurements.

13.
arXiv (CS.CV) 2026-06-11

Plan-and-Verify Video Reward Reasoning with Spatio-Temporal Scene Graph Grounding

Reward models for text-to-video (T2V) generation guide post-training but often fail at fine-grained semantic alignment. We trace this to two structural weaknesses in existing reasoning-based reward models: they do not systematically verify every condition described in the prompt, and the visual evidence supporting each judgment remains implicit in their free-form reasoning. We propose SG-PVR, a video reward model that addresses these limitations through plan-and-verify reasoning grounded in spatio-temporal scene graphs. The verification plan decomposes the prompt into atomic claims, ensuring every requirement is checked. The spatio-temporal scene graph, encoding entities, attributes, and temporally-grounded relations, is extracted from the video and maintained as a persistent structured visual reference throughout reasoning. Each claim is verified against both the video and the scene graph, anchoring judgments in explicit visual evidence. SG-PVR achieves strong performance on semantic alignment, including fine-grained temporal semantics. As a test-time reranker, it further enhances compositional alignment in T2V generation.

14.
arXiv (CS.CL) 2026-06-24

Leveraging Social Media Data for COVID-19 Studies

Nowadays, social media networks have become widely preferred sources of information. Especially during the time of the Coronavirus disease 2019 COVID 19 pandemic, social media has been one of the most used platforms to get the latest news and information related to COVID 19. Social media are popular because they offer free access to their registered users and allow them to do posting, disseminate information, and respond to others postings. With almost 4.6 billion social media users worldwide, it is not surprising the significant amount of information shared through these platforms could affect how people perceive and cope with the pandemic that we are facing right now. With decent use, social media can be a beneficial digital tool to spread reliable news and public awareness for patients, clinicians, and society. Specifically, this chapter describes linguistic, visual, and emotional indicators expressed in user disclosures. Thus, in this chapter, the related studies of social media platforms usage during the COVID 19 pandemic are explored and discussed in detail. This chapter also categorizes social media data used, introduces different deployed machine learning, feature engineering, natural language processing, and survey methods, and outlines directions for future research.

15.
arXiv (CS.LG) 2026-06-24

Scalable Physics-Inspired Transformers for Spin Glasses

arXiv:2606.22984v2 Announce Type: replace-cross Abstract: Efficient sampling of the Boltzmann distribution in frustrated spin glasses is central to statistical mechanics and combinatorial optimization. Despite advances in machine-learning-based approaches, two issues persist: limited understanding of why variational models fail to benefit from increased scale, unlike the monotonic scaling law of large language models; and high computational cost on large systems that negates advantages over classical sampling methods. Here, we develop a physics-inspired transformer with interpretable sparse attention and spin-tailored positional embeddings to address these challenges. By further leveraging FlashAttention for parallel ancestral sampling, it achieves up to two orders of magnitude speedup over vanilla variational autoregressive networks, enabling neural-network simulations of spin-glass systems to unprecedented sizes on a single GPU. It can resolve full probability distributions, free energies, and overlap statistics across temperatures, for Sherrington-Kirkpatrick and 2D or 3D Edwards-Anderson models, where existing machine-learning methods encounter limitations at certain temperatures. This framework thus establishes a scalable paradigm for frustrated spin-glass systems.

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

InstantForget: Update-Free Backdoor Unlearning with Inference-Time Feature Reset

作者:

arXiv:2606.15730v1 Announce Type: cross Abstract: Backdoor unlearning aims to remove a malicious trigger behavior from a deployed model while preserving clean utility. We study the update-free inference-time setting, where model parameters remain frozen. First, we audit a common projection assumption under oracle paired clean and triggered features. Projection succeeds mainly on BadNets and leaves WaNet, Blended, and SIG at 0.683, 0.888, and 0.941 ASR on CIFAR-10 ResNet-18. This failure is not explained by spectral compactness, spatial locality, or subspace misalignment. It is predicted by a logit-triplet gap involving the target margin, target-logit drop, and non-target logit rise. We then introduce InstantForget, a clean-calibrated gated reset that flags anomalous features with a Mahalanobis score and moves only flagged features toward a neutral non-target representation. With one fixed operating point selected on held-out triggered validation, InstantForget reduces average ASR to 0.071 across four non-adaptive CIFAR-10 triggers without triggered samples or parameter updates at deployment. It also reaches 0.981 detection AUROC and transfers to six of eight tested backbones. Reported failures under WaNet, ModelNet10 point blend, two backbone geometries, and adaptive feature-compactness attacks define the method's scope.

17.
arXiv (quant-ph) 2026-06-16

Arbitrarily Configurable Wavefunctions via Imaginary Gauge Phase Imprint in Non-Hermitian Lattices

arXiv:2603.28153v2 Announce Type: replace-cross Abstract: We propose a general framework, termed the imaginary gauge phase imprint (IGPI), which enables engineering arbitrarily configurable wavefunctions with exact solutions and self-organization dynamics in any-dimensional non-Hermitian lattices under imaginary gauge fields. Using this method, we uncover a novel phase with exact critical wavefunctions, dubbed the skin critical phase (SCP), which is marked by unconventional localization, topological-skin, and dynamical characteristics. Furthermore, we validate the IGPI by imprinting and visualizing complex fractal states with Sierpinski-carpet and Koch-snowflake profiles, as well as exotic super-moire and 3D-moire states in regular lattices. Our work not only offers fresh insights into non-Hermitian critical and fractal physics, but also provides a rigorous paradigm for controlling and visualizing wavefunction patterns using the IGPI in engineered non-Hermitian systems.

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

Practical Anonymous Two-Party Gradient Boosting Decision Tree

arXiv:2605.26903v2 Announce Type: replace-cross Abstract: Structured data is well handled by gradient-boosted decision trees (GBDT), which are usually trained on vertically partitioned features across mutually distrustful parties. High speed and interpretability make GBDTs popular in finance and healthcare, where neural networks may fall short. Enabling secure computation for GBDTs poses unique challenges, requiring secure record alignment for comparison. Relying on private set intersection (PSI) is a de facto approach. Mistaking PSI for a safety measure actually exposes which record identifiers (IDs) are shared between the datasets. Although circuit-PSI could help, it is costly for generic uses. New ideas are needed to efficiently train in a "dark forest". Aiming to hide the IDs, we initiate the study of anonymous GBDT training on split data held by two parties. Dual circuit-PSI in our design lets the parties alternate as receiver to run pick-then-sum over local features. Via oblivious programmable pseudorandom functions, we propagate circuit-PSI outputs as shared state across runs. Avoiding universal alignment, we resolve the neglected dilemma that ID hiding incurs a cost that scales with domain size. Next, we halve the cost of ciphertext packing used to convert single-instruction multiple-data homomorphic encryption from (ring) learning with errors in prior secure GBDT (Usenix Security' 23) and related secure machine-learning computations. Comparative experiments show our protocol remains competitive with leaky approaches in efficiency. Enabling ID-hiding aggregation, our techniques can extend to other vertically partitioned analytics.

19.
arXiv (CS.AI) 2026-06-11

Market Design for AI: Beyond the Copyright Binary

arXiv:2606.12260v1 Announce Type: cross Abstract: How can we design a market of human-generated content for use in training AI models that both enables technological progress and preserves individual incentives for high-quality content creation? Existing approaches take polar positions: a "free-for-all" model based on fair use and a "strong intellectual property rights" model. We show that both fail: Free-for-all does not compensate creators, and – by modeling as a static Stackelberg game – strong intellectual property rights also underpower creative incentives. We find this especially true for more innovative creators, a phenomenon we term the "originality penalty." Extending this insight to a dynamic model, we find another market failure undermining AI model performance, even for an initially good model: Such a model induces greater reliance by humans on AI-assisted creation, resulting in homogenized content feeding back into training, which degrades the model performance – a "curse of precision." We further propose a market design with a data intermediary internalizing cross-creator externalities and subsidizing innovative contributions, thereby restoring efficiency.

20.
arXiv (math.PR) 2026-06-16

Risk or Replace: Efficient Asymptotics for Data-Driven Maintenance

arXiv:2606.14706v1 Announce Type: cross Abstract: Condition-based maintenance (CBM) is an approach that plans interventions for deteriorating systems according to their observed operational state. CBM reduces unplanned downtime and extends usable lifetime. We study a heterogeneous population of components that degrade over time according to a stochastic processes with non-negative and i.i.d. increments that are characterized by component-specific parameters that remain unobservable to the decision maker. We rely on degradation data to estimate these parameters and determine replacement actions at equidistant epochs. The goal is to minimize the long-run average cost, which incorporates fixed replacement costs, failure costs, and operating costs. This problem can be formulated as a high-dimensional partially observable Markov decision process (POMDP), which is generally intractable. We develop a tractable, data-driven CBM policy that estimates the optimal policy of a hypothetical Oracle that has full information of the underlying degradation parameters and call this policy the Estimated Oracle's Optimal Policy (EOP). We introduce a scaling regime where both the failure thresholds and cost parameters increase proportionally, reflecting practical settings in which component lifetimes and maintenance costs are large relative to the time between two consecutive CBM decision moments. We show that the regret of the EOP, defined as the difference between its long-run average cost and that of the Oracle, converges to zero in the scaling regime when the parameter estimator is consistent. Across extensive experiments using both real and simulated data, the EOP achieves very low regret and, whenever the optimal POMDP policy can be computed exactly, a negligible optimality gap.

21.
arXiv (quant-ph) 2026-06-16

How Many Shots Are Enough for a Quantum Circuit?

arXiv:2606.16965v1 Announce Type: new Abstract: Quantum algorithms require repeated circuit executions, known as shots, to estimate output distributions accurately. Determining the minimal number of shots needed to meet a target accuracy is crucial to reduce costs and resource usage, especially on today's noisy and expensive quantum hardware. In this paper, we address the shot optimisation problem in a black-box setting, where no assumptions are made about the structure of the quantum circuit or the noise model of the backend. We introduce IncrementalExecution, a novel online framework that dynamically determines when to stop executing shots based on the principle of point of diminishing returns: the point at which additional shots no longer significantly alter the empirical distribution of a fixed circuit. The framework supports customisable policies for shot management, enabling flexible trade-offs between execution cost and result fidelity within static execution scenarios. We assess our proposal through an extensive experimental evaluation spanning 33,750 framework configurations across 180 unique static quantum circuit-backend combinations, for a total of 7.3M independent experiments. Unlike prior work that relies on problem-specific knowledge or algorithm-dependent assumptions (e.g., variational or adaptive workflows), our approach is applicable to a large set of static circuits and immediately deployable on current quantum cloud platforms.

22.
arXiv (CS.LG) 2026-06-15

Recursively Trained Diffusion Models: Limiting Collapse Distribution and Spectral Characterization

arXiv:2606.13796v1 Announce Type: cross Abstract: Recursive training of generative models on their own outputs can lead to model collapse, a compounding drift away from the true data distribution. Existing theoretical works bound finite-round error accumulation in the context of diffusion models, but two questions remain open:~what distribution does the recursion converge to, and how fast? We answer both, isolating a mechanism distinct from imperfect learning: even with perfect score estimation and exact sampling, the early stopping of the reverse diffusion (required for numerical stability) drives a progressive drift away from the data distribution. We prove that this recursion converges geometrically to a unique limiting distribution, which admits a closed-form characterization as an infinite mixture of increasingly Gaussian-smoothed versions of the data distribution. A Hermite spectral decomposition of this limit reveals that recursive training acts as a low-pass filter: higher-order modes, which encode fine non-Gaussian structure, are attenuated much more strongly than coarse modes. This spectral picture motivates annealed truncation schedules that progressively shrink truncation times across retraining rounds; we prove that any schedule converging to $0$ asymptotically eliminates recursive compounding. Finally, we show our idealized characterization is robust: in the presence of discretization and score estimation errors, the learned distribution remains in a Wasserstein-2 ball around the ideal limit, with mode-dependent contraction rates that contract high-order errors faster than low-order ones. We validate the theory on synthetic Gaussian mixtures and CIFAR-10.

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

Efficiently Linking Real Scenes with Synthetic Data Generation for AI-based Cognitive Robotics and Computer Vision Applications

AI vision models are a driving factor for the potential use case scenarios of cognitive robotics within in the industry and household applications. A large array of methods from semantic environment analysis towards 6D and grasping pose estimation have been proposed based on the latest AI achievements. However, such advancements require further strong and efficient methods w.r.t. training data and AI-architectures, which are capable in synergy to tackle current challenges, precision limits, and scalability beyond domain gaps. In this paper, we discuss these current limits and trends in the related state-of-the-art which are challenging those. Further we discuss our current work in progress on bridging the domain gap between simulations and real world applications by linking those in the training data generation.

24.
arXiv (CS.CL) 2026-06-16

Dual-Uncertainty Guided Policy Learning for Multimodal Reasoning

Reinforcement learning with verifiable rewards (RLVR) has advanced reasoning capabilities in multimodal large language models. However, existing methods typically treat visual inputs as deterministic, overlooking the perceptual ambiguity inherent to the visual modality. Consequently, they fail to distinguish whether a model's uncertainty stems from complex reasoning or ambiguous perception, preventing the targeted allocation of exploration or learning signals. To address this gap, we introduce DUPL, a dual-uncertainty guided policy learning approach for multimodal RLVR that quantifies and leverages both perceptual uncertainty (via symmetric KL divergence) and output uncertainty (via policy entropy) to guide policy updates. By establishing an uncertainty-driven feedback loop and employing a dynamic branch prioritization mechanism, DUPL recalibrates the policy advantage to focus learning on states with high perceptual or decisional ambiguity, enabling effective targeted exploration beyond passive data augmentation. Evaluated on diverse multimodal reasoning benchmarks spanning mathematical and general domains, DUPL achieves solid gains. It improves Qwen2.5-VL accuracy by up to $12.3%$ (3B) and $7.9%$ (7B), and Qwen3-VL-Instruct by up to $10.7%$ (4B) and $12.4%$ (8B), consistently outperforming GRPO, while seamlessly generalizing to alternative algorithms (DAPO, $+6.5%$ avg) and architectures (LLaVA-OneVision-1.5, $+4.7%$ avg). These results demonstrate that DUPL is an effective and generalizable approach for multimodal RLVR.

25.
arXiv (CS.CV) 2026-06-15

Memento: Reconstruct to Remember for Consistent Long Video Generation

Long-form video generation requires recurring subjects to remain consistent across various shots, viewpoints, motions, and scene transitions. Existing temporal decomposition methods improve scalability by generating videos shot by shot. However, they mainly focus on optimizing plausible next-shot continuations without verifying whether the historical memory preserves identity-critical subject evidence. Consequently, as generation proceeds, recurring subjects may be diluted, overwritten, or forgotten. In this paper, we propose Memento, a subject-reconstruction-guided framework that treats subject preservation as an explicit identity grounding problem, based on the premise that a memory bank faithfully preserving a subject should support reconstructing that subject from memory alone. Specifically, Memento jointly trains autoregressive next-shot generation with memory-based subject reconstruction, recovering target appearances using historical memory and global story captions. To disentangle long-range subject evidence from short-range cues, Memento introduces a dual-query memory mechanism, where one query retrieves identity-relevant memory and the other selects short-context keyframes for coherent continuation. Additionally, a subject-aware cinematic data pipeline provides precise reconstruction supervision via consistent, pronoun-free subject descriptions. Experiments demonstrate that Memento achieves state-of-the-art performance in long-term subject consistency, cross-shot coherence, and visual quality.