arXiv subjects

Hao Jiang

Publications and source records attributed to Hao Jiang.

At least 91 records · Page 5Linked to original sources

arXiv2026

Recovering Policy-Induced Errors: Benchmarking and Trajectory Synthesis for Robust GUI Agents

While GUI agents have advanced rapidly, they often lack the robustness to recover from their own errors, hindering real-world deployment. To bridge this gap at both the evaluation and data levels, we introduce GUI-RobustEval and propose Robustness-driven Trajectory Synthesis. GUI-RobustEval contains $1,216$ executable test cases that systematically measure error recovery capabilities across a broad and realistic spectrum of error modes. At the data level, RoTS is a scalable synthesis framework that creates $800k$ high-quality data via a tree-based pipeline that proactively discovers diverse error modes and synthesizes corresponding recovery steps. Our two models, RoTS-7B and RoTS-32B, fine-tuned on our dataset, both demonstrate significant gains on GUI-RobustEval and traditional GUI benchmarks. Notably, RoTS-32B achieves state-of-the-art performance on OSWorld, with a $47.4\%$ success rate and a $33.8\%$ All-Pass@4 score, suggesting that improved long-horizon error recovery ability contributes to both robustness and overall performance. Our code is available at https://github.com/AlibabaResearch/RoTS.

cs.CV↗

arXiv2026

Long Live The Balance: Information Bottleneck Driven Tree-based Policy Optimization

Recent advances in online reinforcement learning (RL) for large language models (LLMs) have demonstrated promising performance in complex reasoning tasks. However, they often exhibit an imbalanced exploration-exploitation trade-off, resulting in unstable optimization and sub-optimal performance. We introduce IB-Score, a novel metric grounded in Information Bottleneck theory that evaluates policy's exploration-exploitation balance by quantifying the trade-off between step-level reasoning diversity and mutual information shared with the correct answer. Analysis based on IB-Score shows that popular online RL approaches (e.g., GRPO) with common regularizers fail to consistently maintain balance during training with suboptimal results. To address this, we propose Information Bottleneck-driven Tree-based Policy Optimization (IB-TPO), a principled framework that formulates IB-Score as a fine-grained optimization objective and utilizes a novel IB-guided tree sampling strategy that not only improves the efficiency of online sampling with 50% more trajectories under the same token budget, but also reuses the tree structure for effective IB-Score Monte Carlo estimation. Extensive experiments across standard benchmarks show that our method significantly outperforms GRPO baseline by 2.9% to 3.6% and also outperforms other state-of-the-art online RL approaches. Our code is available at https://github.com/alibaba/EfficientRL.

cs.LG↗

arXiv2026

DynFrame: Adaptive Reasoning-Driven Multimodal Framework with Dynamic Frame Augmentation for Complex Video Understanding

Recent video multimodal large language models (MLLMs) increasingly couple step-by-step reasoning with on-demand visual evidence retrieval, allowing models to revisit relevant video segments during inference. However, two structural gaps remain in existing thinking-with-video systems. (i) Sampling density is not a learnable decision: existing methods may let the model decide where to look, but the per-window frame rate is largely fixed. As a result, fine-grained evidence is often recovered through repeated retrieval calls, which increases inference context length and training difficulty. (ii) Retrieval and answer generation are usually optimized with a single trajectory-level advantage, so the "where to look" tokens and the "how to answer" tokens receive the same credit even when one is correct and the other is not. To address these gaps, we present DynFrame, a framework that emits the temporal window and the sampling density as native tokens within a single autoregressive pass. This learnable span-density retrieval enables acquiring multi-granularity evidence with a single retrieval step. Based on the above tokenized retrieval interface, we further introduce Segment-Decoupled GRPO (SD-GRPO), which splits each rollout at the retrieval boundary and assigns role-specific token-level advantages, separately crediting the sampling decision and the answer. Trained on the curated DM-CoT-74k and DM-RL-45k, DynFrame-4B is competitive with strong 7B-8B baselines across six benchmarks (NExT-GQA, Charades-STA, ActivityNet-MR, Video-MME, MLVU, LVBench), and DynFrame-8B sets new state-of-the-art on most metrics. Code is available at https://github.com/zhangguanghao523/DynFrame.

cs.CV↗

arXiv2026

Geometry-Structured Channel Reconstruction for Conventional and Fluid Antenna Systems: Bayesian Inference and Fundamental Limits

Accurate channel state information (CSI) acquisition is critical for exploiting the spatial flexibility of fluid antenna systems (FASs). However, port selection and transmission optimization require CSI over a large number of candidate port positions, making direct port-wise estimation prohibitively costly in terms of pilot overhead. This paper addresses this challenge through geometry-structured channel reconstruction, which exploits the fact that the port-domain CSI can be parameterized by a small number of dominant propagation paths. We first establish fundamental mean square error (MSE) and normalized MSE (NMSE) benchmarks for both geometry-structured and unstructured channel reconstruction, providing analytical references for evaluating the intrinsic benefit of geometric modeling in conventional antenna systems and FASs. Motivated by the strong spatial correlation induced by densely distributed fluid antenna ports, we further propose a Bayesian reconstruction framework, termed geometry-structured expectation-maximization approximate message passing (GS-EM-AMP). The proposed algorithm incorporates geometric channel structure into the EM-AMP procedure and adaptively learns unknown statistical parameters from noisy observations. Numerical results demonstrate that GS-EM-AMP achieves near-bound reconstruction accuracy while maintaining strong robustness against steering-domain correlation, thereby offering an efficient and reliable solution for large-scale CSI acquisition in FASs.

eess.SP↗

arXiv2026

Neighbor-Consistent Neural Filters for Robust Personal Sound Zones Under Localization Uncertainty

Coordinate-conditioned neural networks can generate head-tracked personal sound zone (PSZ) loudspeaker filters in real time, but they are sensitive to localization uncertainty. Small fluctuations in estimated listener coordinates, caused by optical distortion, temporary occlusions, or tracking jitter, may produce large filter changes even when listeners are physically stationary. This paper proposes neighbor-consistent neural filters that regularize the coordinate-to-filter mapping by penalizing filter differences at randomly perturbed neighboring coordinates during training. To evaluate robustness against tracking noise, we introduce a decoupled protocol that fixes the acoustic transfer functions at a physical anchor while perturbing only the coordinate inputs used for filter generation. Isolation quality and local stability are evaluated using neighborhood median and lower-tail statistics of inter-zone and inter-program isolation, together with spatial variation rates that quantify metric sensitivity within a coordinate neighborhood. In simulation with a split-band woofer-tweeter system and 25 randomly sampled anchor positions, neighbor consistency reduces the root-mean-square (RMS) variation rate by up to 55.9% in the woofer band and 30.3% in the tweeter band while largely preserving isolation quality and improving lower-tail robustness. In in-situ measurements using a 24-driver array and two stationary head-and-torso simulators, the proposed regularization improves worst-case neighborhood isolation by up to 16.9% and reduces spatial variation rates by up to 61.8%. These results demonstrate that neighbor-consistency regularization effectively stabilizes PSZ rendering under localization uncertainty.

eess.AS↗

arXiv2026

Finite-Aperture Planar Fluid Antenna Array

Fluid antenna systems (FASs) are emerging as a reconfigurable-aperture technology that expands physical-layer design beyond fixed, rigid antenna geometries. While the \emph{fading diversity} of FASs -- which exploits spatial channel fluctuations for signal enhancement and interference avoidance -- has been widely studied, the \emph{geometry diversity} created by reconfigurable port placement remains far less understood, particularly for planar architectures under finite-aperture constraints. This paper develops a systematic analytical framework for finite-aperture planar fluid antenna arrays (FAAs). First, we derive a closed-form characterization of the minimum inter-port distance under uniform random placement over a rectangular aperture and show that it follows a Rayleigh law. Its mean scales as $\mathcal{O}(M^{-1})$, in sharp contrast to the $\mathcal{O}(M^{-2})$ behavior in the linear case in which $M$ represents the number of candidate ports, revealing a fundamentally more favorable packing geometry in two dimensions. Secondly, we establish a universal Cramér-Rao bound (CRB) for joint elevation-azimuth estimation, governed by a $2\times 2$ \emph{geometric inertia matrix} whose determinant and eigenstructure fully capture the role of port placement in estimation precision. We further prove that both the trace and determinant of this matrix are invariant to the azimuth look direction. Third, we uncover an intrinsic \emph{precision--ambiguity trade-off}: maximizing the geometric determinant to minimize the CRB drives ports toward the aperture boundary, but simultaneously increases sidelobe-induced spatial ambiguity.

cs.IT↗

arXiv2026

Structured Labeling Enables Faster Vision-Language Models for End-to-End Autonomous Driving

Vision-Language Models (VLMs) offer a promising approach to end-to-end autonomous driving due to their human-like reasoning capabilities. However, troublesome gaps remains between current VLMs and real-world autonomous driving applications. One major limitation is that existing datasets with loosely formatted language descriptions are not machine-friendly and may introduce redundancy. Additionally, high computational cost and massive scale of VLMs hinder the inference speed and real-world deployment. To bridge the gap, this paper introduces a structured and concise benchmark dataset, NuScenes-S, which is derived from the NuScenes dataset and contains machine-friendly structured representations. Moreover, we present FastDrive, a compact VLM baseline with 0.9B parameters. In contrast to existing VLMs with over 7B parameters and unstructured language processing(e.g., LLaVA-1.5), FastDrive understands structured and concise descriptions and generates machine-friendly driving decisions with high efficiency. Extensive experiments show that FastDrive achieves competitive performance on structured dataset, with approximately 20% accuracy improvement on decision-making tasks, while surpassing massive parameter baseline in inference speed with over 10x speedup. Additionally, ablation studies further focus on the impact of scene annotations (e.g., weather, time of day) on decision-making tasks, demonstrating their importance on decision-making tasks in autonomous driving.

cs.CV↗

arXiv2026

Think When Needed: Adaptive Reasoning-Driven Multimodal Embeddings with a Dual-LoRA Architecture

Multimodal large language models (MLLMs) have emerged as a powerful backbone for multimodal embeddings. Recent methods introduce chain-of-thought (CoT) reasoning into the embedding pipeline to improve retrieval quality, but remain costly in both model size and inference cost. They typically employ separate reasoner and embedder with substantial parameter overhead, and generate CoT indiscriminately for every input. However, we observe that for simple inputs, discriminative embeddings already perform well, and redundant reasoning can even mislead the model, degrading performance. To address these limitations, we propose Think When Needed (TWN), a unified multimodal embedding framework with adaptive reasoning. TWN introduces a dual-LoRA architecture that attaches reasoning and embedding adapters to a shared frozen backbone, detaching gradients at their interface to mitigate gradient conflicts introduced by joint optimization while keeping parameters close to a single model. Building on this, an adaptive think mechanism uses a self-supervised routing gate to decide per input whether to generate CoT, skipping unnecessary reasoning to reduce inference overhead and even improve retrieval quality. We further explore embedding-guided RL to optimize CoT quality beyond supervised training. On the 78 tasks of MMEB-V2, TWN achieves state-of-the-art embedding quality while being substantially more efficient than existing generative methods, requiring only 3-5% additional parameters relative to the backbone and up to 50% fewer reasoning tokens compared to the full generative mode.

cs.CV↗

arXiv2026

Joint Segment Activation and Antenna Placement for Uplink SWAN Systems

This article analyzes the achievable sum-rate of multiuser uplink segmented waveguide-enabled pinching-antenna systems (SWANs). To unveil system-design insights, an upper bound on the achievable sum-rate is derived, based on which the existence of an optimal segment activation level is theoretically established. Motivated by this result, hybrid segment selection and aggregation (HSS/A) schemes are proposed to jointly optimize segment activation and pinching-antenna (PA) placement. Correspondingly, low-complexity greedy algorithms are developed for the considered optimization problem. Numerical results validate the theoretical analysis and demonstrate that the proposed HSS/A schemes outperform conventional full-segment aggregation.

eess.SP↗

arXiv2026

Muninn: Your Trajectory Diffusion Model But Faster

Diffusion-based trajectory planners can synthesize rich, multimodal robot motions, but their iterative denoising makes online planning and control prohibitively slow. Existing accelerations either modify the sampler or compress the network--sacrificing plan quality or requiring retraining without accounting for downstream control risk. We address the problem of making diffusion-based trajectory planners fast enough for real-time robot use without retraining the model or sacrificing trajectory quality, and in a way that works across diverse state-space diffusion architectures. Our key insight is that diffusion trajectory planners expose two signals we can exploit: a cheap probe of how their internal trajectory representation changes across steps, and analytic coefficients that describe how denoiser errors affect the sampler's state update. By calibrating the first signal against the second on offline runs, we obtain a per-step score that upper-bounds how far the final trajectory can deviate when we reuse a cached denoiser output, and we treat this bound as an uncertainty budget that we can spend over the denoising process. Building on this insight, we present Muninn, a training-free caching wrapper that tracks this uncertainty budget during sampling and, at each diffusion step, chooses between reusing a cached denoiser output when the predicted deviation is small and recomputing the denoiser when it is not. Across standard benchmarks Muninn delivers up to 4.6x wall-clock speedups across several trajectory diffusion models by reducing denoiser evaluations, while preserving task performance and safety metrics. Muninn further certifies that cached rollouts remain within a specified distance of their full-compute counterparts, and we validate these gains in real-time closed-loop navigation and manipulation hardware deployments. Project page: https://github.com/gokulp01/Muninn.

cs.RO↗

arXiv2026

Exposing and Mitigating Temporal Attack in Deepfake Video Detection

While spatiotemporal deepfake detectors achieve high AUC, our experiments reveal their susceptibility to evasion attacks. These models tend to overfit on fragile temporal spectrum cues, rather than learning robust semantic causality. To mitigate this vulnerability, we propose SpInShield, a temporal spectral-invariant defense framework explicitly designed to decouple semantic motion from manipulatable spectral artifacts. We propose a learnable spectral adversary that dynamically synthesizes severe spectral deformations, simulating extreme attack scenarios. By employing a shortcut suppression optimization strategy, SpInShield compels the encoder to extract reliable forensic cues while purging unstable spectral statistics from the latent space. Experiments show that SpInShield obtains competitive performance on widely used datasets and outperforms the strongest baseline by 21.30 percentage points in AUC under simulated amplitude spectral attacks.

cs.CV↗

arXiv2026

Diffusion-APO: Trajectory-Aware Direct Preference Alignment for Video Diffusion Transformers

Efficiently aligning large-scale video diffusion models with human intent requires a scalable and trajectory-aware pathway that bridges the inherent discrepancy between training noise distributions and practical inference trajectories. While existing paradigms such as Direct Preference Optimization (DPO) and Group Relative Policy Optimization (GRPO) attempt to address this, they are often hindered by either reliance on bias-prone, complex reward models or suboptimal timestep sampling. In this paper, we propose Diffusion-APO (Aligned Preference Optimization), a trajectory-aware algorithm that resolves this misalignment by synchronizing training noise with inference-time denoising paths to maximize gradient signal efficacy. To translate this algorithmic innovation into a practical solution, we introduce a unified and modular RLHF framework that integrates online ranking, half-online anchoring, offline refinement, and distillation-aware drift correction. This framework enables flexible, multi-stage preference alignment across diverse data and computational constraints without relying on scalar-reward-based policy gradients. Through extensive experiments, we demonstrate that Diffusion-APO consistently outperforms standard baselines in visual quality and instruction following, while effectively preserving generative fidelity during model acceleration, providing a robust, end-to-end pathway for scalable video diffusion alignment.

cs.CV↗

arXiv2026

Over-the-Air Computation via Segmented Waveguide-Enabled Pinching-Antenna Systems

A segmented waveguide-enabled pinching-antenna system (SWAN)-assisted over-the-air computation (AirComp) framework is proposed. Three transmission architectures, namely segment selection (SS), phase-shifter-free segment aggregation (SA), and phase-shifter-enabled SA, are developed for uplink signal aggregation. For each architecture, low-complexity algorithms are developed to optimize the pinching-antenna placement and the per-segment phase shifts. Numerical results demonstrate the effectiveness of the proposed approaches and the superiority of SWAN over the conventional pinching-antenna system (PASS). It is shown that both SS and SA achieve lower computation mean-squared error than the conventional PASS, while segment-wise phase control further improves the performance of SA.

eess.SP↗

arXiv2026

Towards Customized Multimodal Role-Play

Unified multimodal understanding and generation models enable richer human-AI interaction. Yet jointly customizing a character's persona, dialogue style, and visual identity while maintaining output consistency across modalities remains largely unexplored. To mitigate this gap, we introduce a new task, Customized Multimodal Role-Play (CMRP). We construct the RoleScape-20 dataset comprising 20 characters, including training and evaluation data that cover persona, stylistic descriptions, visual/expressive cues, and text-image interactions. Building on a unified model, we devise UniCharacter, a two-stage training framework containing Unified Supervised Finetuning (Unified-SFT) and character-specific group relative policy optimization (Character-GRPO). Given only 10 images plus corresponding interaction examples, the model acquires the target character and exhibits coherent persona, style, and visual identity in both generated text and images. This process takes about 100 GPU hours. Experiments on the RoleScape-20 dataset show that the proposed method substantially outperforms prior approaches. Ablation studies further validate the effectiveness of our cross-modal consistency design and few-shot customization strategy. We argue that CMRP, coupled with unified modeling, provides a basis for next-generation characterful and immersive interactive agents.

cs.LG↗

arXiv2026

PromptEcho: Annotation-Free Reward from Vision-Language Models for Text-to-Image Reinforcement Learning

Reinforcement learning (RL) can improve the prompt following capability of text-to-image (T2I) models, yet obtaining high-quality reward signals remains challenging: CLIP Score is too coarse-grained, while VLM-based reward models (e.g., RewardDance) require costly human-annotated preference data and additional fine-tuning. We propose PromptEcho, a reward construction method that requires \emph{no} annotation and \emph{no} reward model training. Given a generated image and a guiding query, PromptEcho computes the token-level cross-entropy loss of a frozen VLM with the original prompt as the label, directly extracting the image-text alignment knowledge encoded during VLM pretraining. The reward is deterministic, computationally efficient, and improves automatically as stronger open-source VLMs become available. For evaluation, we develop DenseAlignBench, a benchmark of concept-rich dense captions for rigorously testing prompt following capability. Experimental results on two state-of-the-art T2I models (Z-Image and QwenImage-2512) demonstrate that PromptEcho achieves substantial improvements on DenseAlignBench (+26.8pp / +16.2pp net win rate), along with consistent gains on GenEval, DPG-Bench, and TIIFBench without any task-specific training. Ablation studies confirm that PromptEcho comprehensively outperforms inference-based scoring with the same VLM, and that reward quality scales with VLM size. We will open-source the trained models and the DenseAlignBench.

cs.CV↗

arXiv2026

Cramer-Rao Bounds for Activity Detection in Conventional and Fluid Antenna Systems

In this letter, we develop a unified Cramér-Rao bound (CRB) framework to characterize the fundamental performance limits of transmission activity detection in fluid antenna systems (FASs) and conventional multiple fixed-position antenna (FPA) systems. To facilitate CRB analysis applicable to activity indicators, we relax the binary activity states to continuous parameters, thereby aligning the bound-based evaluation with practical threshold-based detection decisions. Closed-form CRB expressions are derived for two representative detection formulations, namely covariance-oriented and coherent models. Moreover, for single-antenna FASs, we obtain a closed-form coherent CRB by leveraging random matrix theory. The results demonstrate that CRB-based analysis provides a tractable and informative benchmark for evaluating activity detection across architectures and detection schemes, and further reveal that FASs can deliver strong spatial-diversity gains with significantly reduced complexity.

cs.IT↗

arXiv2026

ReST: A Plug-and-Play Spatially-Constrained Representation Enhancement Framework for Local-Life Recommendation

Local-life recommendation have witnessed rapid growth, providing users with convenient access to daily essentials. However, this domain faces two key challenges: (1) spatial constraints, driven by the requirements of the local-life scenario, where items are usually shown only to users within a limited geographic area, indirectly reducing their exposure probability; and (2) long-tail sparsity, where few popular items dominate user interactions, while many high-quality long-tail items are largely overlooked due to imbalanced interaction opportunities. Existing methods typically adopt a user-centric perspective, such as modeling spatial user preferences or enhancing long-tail representations with collaborative filtering signals. However, we argue that an item-centric perspective is more suitable for this domain, focusing on enhancing long-tail items representation that align with the spatially-constrained characteristics of local lifestyle services. To tackle this issue, we propose ReST, a Plug-And-Play Spatially-Constrained Representation Enhancement Framework for Long-Tail Local-Life Recommendation. Specifically, we first introduce a Meta ID Warm-up Network, which initializes fundamental ID representations by injecting their basic attribute-level semantic information. Subsequently, we propose a novel Spatially-Constrained ID Representation Enhancement Network (SIDENet) based on contrastive learning, which incorporates two efficient strategies: a spatially-constrained hard sampling strategy and a dynamic representation alignment strategy. This design adaptively identifies weak ID representations based on their attribute-level information during training. It additionally enhances them by capturing latent item relationships within the spatially-constrained characteristics of local lifestyle services, while preserving compatibility with popular items.

cs.IR↗

arXiv2026

Jointly Correlated Dual-Side Fluid Antenna System

Fluid antenna systems (FASs) have introduced a new paradigm for wireless system design by revealing how mutual correlation can be exploited to harvest inherent spatial diversity. While existing studies have mainly focused on one-sided FAS configurations, i.e., with FAS deployed at either the transmitter or the receiver, this work investigates the ergodic capacity of a jointly correlated dual-side FAS under statistical eigenmode transmission. Specifically, a jointly correlated dual-side channel model is developed, and the corresponding ergodic capacity together with a tight closed-form upper bound is derived. In addition, the optimal power allocation is studied, and a practical iterative algorithm is proposed for its implementation.

cs.IT↗