arXiv subjects

Hong Cai

Publications and source records attributed to Hong Cai.

At least 19 recordsLinked to original sources

arXiv2026

TileWeave: Heterogeneous Tiling for Sparse Prefill on Mobile SoCs

Smartphone applications increasingly require large language models to process long contexts, whereas high prefill latency and memory demands constrain local inference. Block-sparse attention reduces computation and memory traffic, but its irregular selections conflict with mobile NPUs' preference for large, regular tiles. Efficient execution must exploit cross-query reuse while avoiding redundant computation and preserving selections needed for accuracy. In this paper, we present TileWeave, which partitions sparse attention by this cross-query reuse: highly shared key blocks run in coarse NPU tiles, while low-reuse blocks run on the GPU only for the query blocks that selected them. In addition, we design an NPU-native pooled-key selector that keeps selection inexpensive, and a runtime that publishes GPU partials in the order the NPU consumes them and merges them in its existing output store. On RULER, HeteroSparse matches XAttention's accuracy with Qwen3-1.7B and improves it with Llama-3.2-3B-Instruct; its mean LongBench scores match XAttention as well. On Snapdragon SM8750 and SM8850 phones, it achieves 1.35--3.23$\times$ faster full-model prefill than XAttention over 1k--16k tokens and up to 1.58$\times$ faster than dense attention.

cs.LG↗

arXiv2026

RankLane: Efficient MoE Prefill on Accelerators with Ranked Expert Capacities

Mixture-of-experts (MoE) models expand model capacity by activating only a subset of experts for each token. On accelerators with static execution plans, expert capacities and placement are fixed before routing reveals each expert's load. Uniform worst-case capacities waste work on padding; smaller capacities tied to expert identities can overflow. We present RankLane, which ranks each device's experts by current load and assigns them to fixed-capacity lanes. Runtime indices select experts and token rows while tier populations, shapes and device placement remain fixed. An aggregation graph preserves device ownership and reduces each token's local contributions on one core. Routing calibration and measured execution costs guide tier selection; overflowing chunks rerun at full capacity. We evaluate full-model prefill of Qwen3-30B-A3B and Gemma 4 26B-A4B with MXFP6 weights on four Cloud AI 100 Ultra SoCs, using 128-2048-token chunks and including overflow detection and in-program recovery support. Against a dense baseline, RankLane achieves end-to-end prefill speedups of 1.27-4.18x on Qwen3 and 1.93-3.46x on Gemma.

cs.LG↗

arXiv2026

Photonic Chips for Universal Quantum Computing: Retrospective and Prospective

Integrated photonic quantum chips have emerged as a promising platform for realizing fault-tolerant photonic quantum computers. Compared with bulk-optics systems, integrated photonics offer compact device footprints, enhanced programmability, and exceptional stability, attributes that are essential for executing large-scale and computationally demanding quantum tasks. In this review, we provide a comprehensive overview of recent advances in both discrete-variable and continuous-variable universal photonic quantum computation. We survey progress in fault-tolerant architectures and state-of-the-art integrated platforms and highlight the key challenges that must be addressed to scale photonic systems toward full fault tolerance. Finally, we outline future opportunities and research directions aimed at bridging the gap between current proof-of-principle demonstrations and fully scalable quantum microprocessors.

quant-ph↗

arXiv2026

Alleviating Hallucination in Reasoning Tasks with Training-Free Uncertainty-Guided Steering

Recent work on hallucination detection in large language models has shown that, for a fixed pre-trained model and reasoning task, it is possible to estimate the model's confidence in the correctness of its outputs. Such uncertainty estimates have primarily been used to improve truthfulness by detecting or filtering confabulations. In this work, we ask whether these signals can instead be used more proactively to directly improve the accuracy of model-generated answers. We propose USteer, a simple, training-free steering mechanism that adjusts a model's layer-wise activations during inference using the gradient of a confidence measure with respect to the activations. This procedure nudges generation toward outputs with lower uncertainty at inference time, without modifying model parameters or requiring additional supervision. We show that this approach consistently reduces hallucination across a range of tasks, demonstrating that confidence signals can be leveraged not only for detection, but also for effective inference-time control of model behavior.

cs.AI↗

arXiv2026

Versatile Quantum Machine Learning with an Ultra-low Power Photonic Quantum Reservoir Computer

Integrated photonic microprocessors provide high-bandwidth, massively parallel linear computation, but realizing nonlinear feature maps and temporal memory remain key challenges for machine learning. Conventional approaches rely on active tuning and additional nonlinear elements, increasing architectural complexity and power overhead. Here we demonstrate an integrated photonic quantum reservoir computer that achieves nonlinear mapping, fading memory, and task versatility without active tuning of the reservoir core. The same chip supports accurate static classification, dynamic prediction, and stable autonomous forecasting, establishing broad utility across both classification and temporal inference tasks. Competitive performance is retained in the zero-bias state, where all on-chip phase shifters are unpowered, eliminating active control and reducing computational power consumption to zero. This passive operation highlights a scalable route to multifunctional machine-learning hardware, where large-scale photonic quantum processors can be repurposed as reservoirs without reconfiguring their internal optical networks. By combining quantum-state encoding, multimode interferometric mixing, and photon-statistical readout, this architecture provides a physically grounded paradigm for low-power, large-scale quantum reservoir computing.

quant-ph↗

arXiv2026

MagnifiQ: Patch-aware Text Guided Progressive Upscaling for High-Resolution Image Restoration

High-resolution image restoration from degraded inputs is challenging because it must preserve global structural consistency while recovering fine-grained local details, especially at 4K resolution where direct diffusion-based restoration is computationally expensive and prone to repeated or inconsistent textures. In this work, we introduce MagnifiQ, an image restoration framework that progressively upscales and restores images across resolutions, e.g., from 1024x1024 to 4096x4096. Our approach leverages a pre-trained text-to-image diffusion model such as SDXL and adapts it for more scalable high-resolution inference by replacing its original self-attention layers with convolutional operations whose computational cost grows linearly with image resolution. We further propose a progressive upscaling strategy that iteratively restores images over multiple resolution stages, refining each intermediate output rather than directly hallucinating the final 4K image, thereby improving global coherence and reducing high-resolution artifacts. To enhance local details while controlling content drift, MagnifiQ uses patch-specific text prompts that provide spatially localized semantic guidance during restoration. Extensive experiments on synthetic and real-world degraded images show that MagnifiQ outperforms prior diffusion-based restoration methods in perceptual quality and human preference, producing sharper textures and more coherent 4K results while offering practical speed--quality trade-offs through its scalable backbone and progressive design.

cs.CV↗

arXiv2026

SNM-VFI: Symmetric Nonlinear Motion-Guided Generative Video Frame Interpolation

We propose Symmetric Nonlinear Motion-guided Generative Video Frame Interpolation (SNM-VFI), a training-free framework for motion-controllable generative video frame interpolation with pre-trained optical flow and video diffusion models. Unlike conventional diffusion-based VFI methods that synthesize intermediate frames from random noise, SNM-VFI guides the generative process with correspondence-aware frames produced by a symmetric nonlinear motion model. Specifically, we first utilize a pre-trained optical flow model to construct multi-frame nonlinear flow-based intermediate frames and confidence maps. These flow-guided frames are then encoded as latent priors to initialize and iteratively guide a pre-trained Video Diffusion model, enabling the diffusion model to preserve dense motion correspondence while improving perceptual realism. To further enhance output quality, we employ confidence maps to fuse structurally reliable flow-based predictions with diffusion-generated details in uncertain regions such as occlusions and object boundaries. Extensive evaluations on challenging benchmarks, including DAVIS, Sintel, and KITTI, demonstrate that SNM-VFI achieves strong perceptual quality, competitive reconstruction accuracy, and robust temporal coherence across diverse motion scenarios.

cs.CV↗

arXiv2026

CLEAR: Closed-Loop Reinforcement Learning at Scale for End-to-End Autonomous Driving

End-to-end autonomous driving (E2E-AD) aims to directly map raw sensor information to driving actions. Recently, with the rapid advancement of multi-modal large language models (MLLMs), researchers have proposed the paradigm of Vision-Language-Action (VLA) models for E2E-AD, where it seeks to integrate visual perception, language understanding and action prediction within a single policy. However, existing VLA-based policies largely adopts imitation learning, where it only learns to drive by optimizing distance-based metrics w.r.t. logged expert trajectories. Such distribution shift between open-loop training and closed-loop inference leads to suboptimal performance in closed-loop planning. To close this gap, we present CLEAR, a system that enables closed-loop training using Reinforcement Learning (RL) at scale for E2E-AD. We propose to learn a novel residual waypoint policy around the waypoint prior from pretrained VLA policies, effectively harnessing the knowledge within. On another front, one of the key challenges to scale up RL for vision-based policies is the number of parallel simulation environments since RL is data hungry. To that end, we design a heterogeneous pipeline that places the simulator and the VLA learner on distinct compute groups, which allows us to dramatically increase the number of simulation environments running in parallel while avoiding resource contention and maintaining training stability. We show that with a simple reward, CLEAR significantly outperforms previous methods and sets new state-of-the-art performance on the challenging benchmarks of CARLA longest6 v2 and Bench2Drive.

cs.RO↗

arXiv2026

SciFlow: Semantic Cross Interference for Self-Supervised Optical Flow Domain Generalization

Motions of objects and scenes carry essential intelligence in video understanding, offering rich cues for interpreting dynamic settings and interactions. Due to the cost and scarcity of high-quality annotation or ground truth of pixel-wise optical flow, however, motion estimation models are typically trained in synthetic domains while deployed in real-world domains. Addressing synthetic-to-real domain generalization challenges has been crucial for developing practical solutions in diverse open-world use cases. This paper introduces SciFlow, a simple yet effective, network-agnostic, training-based approach that leverages self-supervised learning to generalize motion estimation across synthetic and open-world domains. Specifically, SciFlow imposes semantic interference from open-world images onto synthetic images during training, blending indomain features with cross-domain interference, which enables the network to adapt to the real-world domains. Additionally, SciFlow utilizes geometric consistency to ensure validity of the self-supervision. Our experiment results show that SciFlow not only significantly enhances model robustness amidst domain variations, but also remarkably enables synthetic-to-real domain generalization without requiring any ground truth in the open world.

cs.CV↗

arXiv2026

RoCA: Robust Cross-Domain End-to-End Autonomous Driving

End-to-end (E2E) autonomous driving has recently emerged as a new paradigm, offering significant potential. However, few studies have looked into the practical challenge of deployment across domains (e.g., cities). Although several works have incorporated Large Language Models (LLMs) to leverage their open-world knowledge, LLMs do not guarantee cross-domain driving performance and may incur prohibitive retraining costs during domain adaptation. In this paper, we propose RoCA, a novel framework for robust cross-domain E2E autonomous driving. RoCA formulates the joint probabilistic distribution over the tokens that encode ego and surrounding vehicle information in the E2E pipeline. Instantiating with a Gaussian process (GP), RoCA learns a set of basis tokens with corresponding trajectories, which span diverse driving scenarios. Then, given any driving scene, it is able to probabilistically infer the future trajectory. By using RoCA together with a base E2E model in source-domain training, we improve the generalizability of the base model, without requiring extra inference computation. In addition, RoCA enables robust adaptation on new target domains, significantly outperforming direct finetuning. We extensively evaluate RoCA on various cross-domain scenarios and show that it achieves strong domain generalization and adaptation performance.

cs.CV↗

arXiv2026

MAPLE: Latent Multi-Agent Play for End-to-End Autonomous Driving

Vision-language-action (VLA) models are effective as end-to-end motion planners, but can be brittle when evaluated in closed-loop settings due to being trained under traditional imitation learning framework. Existing closed-loop supervision approaches lack scalability and fail to completely model a reactive environment. We propose MAPLE, a novel framework for reactive, multi-agent rollout of a dynamic driving scenario in the latent space of the VLA model. The ego vehicle and nearby traffic agents are independently controlled over multi-step horizons, while being reactive to other agents in the scene, enabling closed-loop training. MAPLE consists of two training stages: (1) supervised fine-tuning on the latent rollouts based on ground-truth trajectories, followed by (2) reinforcement learning with global and agent -specific rewards that encourage safety, progress, and interaction realism. We further propose diversity rewards that encourage the model to generate planning behaviors that may not be present in logged driving data. Notably, our closed-loop training framework is scalable and does not require external simulators, which can be computationally expensive to run and have limited visual fidelity to the real-world. MAPLE achieves state-of-the-art driving performance on Bench2Drive and demonstrates scalable, closed-loop multi-agent play for robust E2E autonomous driving systems.

cs.RO↗

arXiv2026

FALO: Fast and Accurate LiDAR 3D Object Detection on Resource-Constrained Devices

Existing LiDAR 3D object detection methods predominantely rely on sparse convolutions and/or transformers, which can be challenging to run on resource-constrained edge devices, due to irregular memory access patterns and high computational costs. In this paper, we propose FALO, a hardware-friendly approach to LiDAR 3D detection, which offers both state-of-the-art (SOTA) detection accuracy and fast inference speed. More specifically, given the 3D point cloud and after voxelization, FALO first arranges sparse 3D voxels into a 1D sequence based on their coordinates and proximity. The sequence is then processed by our proposed ConvDotMix blocks, consisting of large-kernel convolutions, Hadamard products, and linear layers. ConvDotMix provides sufficient mixing capability in both spatial and embedding dimensions, and introduces higher-order nonlinear interaction among spatial features. Furthermore, when going through the ConvDotMix layers, we introduce implicit grouping, which balances the tensor dimensions for more efficient inference and takes into account the growing receptive field. All these operations are friendly to run on resource-constrained platforms and proposed FALO can readily deploy on compact, embedded devices. Our extensive evaluation on LiDAR 3D detection benchmarks such as nuScenes and Waymo shows that FALO achieves competitive performance. Meanwhile, FALO is 1.6~9.8x faster than the latest SOTA on mobile Graphics Processing Unit (GPU) and mobile Neural Processing Unit (NPU).

cs.CV↗

arXiv2026

CoReDiT: Spatial Coherence-Guided Token Pruning and Reconstruction for Efficient Diffusion Transformers

Diffusion Transformers (DiTs) deliver remarkable image and video generation quality but incur high computational cost, limiting scalability and on-device deployment. We introduce CoReDiT, a structured token pruning framework for DiTs across vision tasks. CoReDiT uses a linear-time spatial coherence score to estimate local redundancy in the latent token lattice and skips high coherence (redundant) tokens in self-attention. To maintain a dense representation and avoid visual discontinuities, we reconstruct skipped attention outputs via coherence-guided aggregation of spatially neighboring retained tokens. We further introduce a progressive, block-adaptive pruning schedule that increases pruning gradually and allocates larger budgets to blocks and denoising steps with higher redundancy. Across state-of-the-art diffusion backbones including PixArt-α and MagicDrive-V2, CoReDiT achieves up to 55% self-attention FLOPs reduction and inference speedups of 1.33x on cloud GPUs and 1.72x on mobile NPUs, while maintaining high visual quality. Notably, CoReDiT also increases on-device memory head-room, enabling higher-resolution generation.

cs.CV↗

arXiv2026

BePo: Dual Representation for 3D Occupancy Prediction

3D occupancy infers fine-grained 3D geometry and semantics which is critical for autonomous driving. Most existing approaches carry high compute costs, requiring dense 3D feature volume and cross-attention to effectively aggregate information. More efficient methods adopt Bird's Eye View (BEV) or sparse points as scene representation leading to much reduced runtime. However, BEV struggles with small objects that often have very limited feature representation especially after being projected to the ground plane. Sparse points on the other and, can model objects of various sizes in 3D space, but is inefficient at capturing flat surfaces or large objects. To address these shortcomings, we present BePo, which features a dual representation of BEV and sparse points. The 3D information learned in the sparse points branch is shared with the BEV stream via cross-attention, which injects learning signals of difficult objects on the BEV plane. The outputs of both branches are then fused to generate the final 3D occupancy predictions. Extensive experiments on a suite of challenging benchmarks including Occ3D-nuScenes, Occ3D-Waymo and Occ-ScanNet demonstrate the superiority of our proposed BePo. In addition, BePo carries low inference cost even when compared to latest efficient methods.

cs.CV↗

arXiv2026

Enhancing Novel View Synthesis via Geometry Grounded Set Diffusion

We present SetDiff, a geometry-grounded multi-view diffusion framework that enhances novel-view renderings produced by 3D Gaussian Splatting. Our method integrates explicit 3D priors, pixel-aligned coordinate maps and pose-aware Plucker ray embeddings, into a set-based diffusion model capable of jointly processing variable numbers of reference and target views. This formulation enables robust occlusion handling, reduces hallucinations under low-signal conditions, and improves photometric fidelity in visual content restoration. A unified set mixer performs global token-level attention across all input views, supporting scalable multi-camera enhancement while maintaining computational efficiency through latent-space supervision and selective decoding. Extensive experiments on EUVS, Para-Lane, nuScenes, and DL3DV demonstrate significant gains in perceptual fidelity, structural similarity, and robustness under severe extrapolation. SetDiff establishes a state-of-the-art diffusion-based solution for realistic and reliable novel-view synthesis in autonomous driving scenarios.

cs.CV↗

arXiv2026

HLA: Hadamard Linear Attention

The attention mechanism is an important reason for the success of transformers. It relies on computing pairwise relations between tokens. To reduce the high computational cost of standard quadratic attention, linear attention has been proposed as an efficient approximation. It employs kernel functions that are applied independently to the inputs before the pairwise similarities are calculated. That allows for an efficient computational procedure which, however, amounts to a low-degree rational function approximating softmax. We propose Hadamard Linear Attention (HLA). Unlike previous works on linear attention, the nonlinearity in HLA is not applied separately to queries and keys, but, analogously to standard softmax attention, after the pairwise similarities have been computed. It will be shown that the proposed nonlinearity amounts to a higher-degree rational function to approximate softmax. An efficient computational scheme for the proposed method is derived that is similar to that of standard linear attention. In contrast to other approaches, no time-consuming tensor reshaping is necessary to apply the proposed algorithm. The effectiveness of the approach is demonstrated by applying it to a large diffusion transformer model for video generation, an application that involves very large amounts of tokens.

cs.AI↗

arXiv2026

Double-P: Hierarchical Top-P Sparse Attention for Long-Context LLMs

As long-context inference becomes central to large language models (LLMs), attention over growing key-value caches emerges as a dominant decoding bottleneck, motivating sparse attention for scalable inference. Fixed-budget top-k sparse attention cannot adapt to heterogeneous attention distributions across heads and layers, whereas top-p sparse attention directly preserves attention mass and provides stronger accuracy guarantees. Existing top-p methods, however, fail to jointly optimize top-p accuracy, selection overhead, and sparse attention cost, which limits their overall efficiency. We present Double-P, a hierarchical sparse attention framework that optimizes all three stages. Double-P first performs coarse-grained top-p estimation at the cluster level using size-weighted centroids, then adaptively refines computation through a second top-p stage that allocates token-level attention only when needed. Across long-context benchmarks, Double-P consistently achieves near-zero accuracy drop, reducing attention computation overhead by up to 1.8x and delivers up to 1.3x end-to-end decoding speedup over state-of-the-art fixed-budget sparse attention methods.

cs.LG↗

arXiv2026

Generative Scenario Rollouts for End-to-End Autonomous Driving

Vision-Language-Action (VLA) models are emerging as highly effective planning models for end-to-end autonomous driving systems. However, current works mostly rely on imitation learning from sparse trajectory annotations and under-utilize their potential as generative models. We propose Generative Scenario Rollouts (GeRo), a plug-and-play framework for VLA models that jointly performs planning and generation of language-grounded future traffic scenes through an autoregressive rollout strategy. First, a VLA model is trained to encode ego vehicle and agent dynamics into latent tokens under supervision from planning, motion, and language tasks, facilitating text-aligned generation. Next, GeRo performs language-conditioned autoregressive generation. Given multi-view images, a scenario description, and ego-action questions, it generates future latent tokens and textual responses to guide long-horizon rollouts. A rollout-consistency loss stabilizes predictions using ground truth or pseudo-labels, mitigating drift and preserving text-action alignment. This design enables GeRo to perform temporally consistent, language-grounded rollouts that support long-horizon reasoning and multi-agent planning. On Bench2Drive, GeRo improves driving score and success rate by +15.7 and +26.2, respectively. By integrating reinforcement learning with generative rollouts, GeRo achieves state-of-the-art closed-loop and open-loop performance, demonstrating strong zero-shot robustness. These results highlight the promise of generative, language-conditioned reasoning as a foundation for safer and more interpretable end-to-end autonomous driving.

cs.CV↗