arXiv subjects

Fan Wang

Publications and source records attributed to Fan Wang.

At least 55 records · Page 3Linked to original sources

arXiv2026

WorldOlympiad: Can Your World Model Survive a Triathlon?

We introduce WorldOlympiad, a benchmark for diagnosing video-based world models across physical faithfulness, geometric consistency, and interaction fidelity. While existing benchmarks often focus on visual quality, semantic alignment, or short-term temporal coherence, they provide limited insight into whether generated videos obey physical rules, preserve coherent 3D structure, and sustain controllable interactions over long horizons. To address this gap, WorldOlympiad decomposes world-model evaluation into three complementary dimensions. The physical track uses object segmentation and MLLM-as-judge to assess whether generated videos follow interpretable rules in mechanics, thermal phenomena, and material properties. The geometry track reconstructs generated videos with Gaussian splatting and evaluates structural consistency, cross-view coherence, and camera-trajectory alignment. The interaction track assesses whether generated rollouts follow complex action prompts and maintain smooth, coherent transitions across consecutive video chunks. WorldOlympiad further covers three major downstream scenarios, including gaming, robotics, and general real-world videos, capturing diverse challenges from interactive control and embodied manipulation to open-domain motion and camera dynamics. Together, these tracks and scenarios form a scalable and interpretable evaluation suite that exposes failure modes beyond generic video quality. Experiments on state-of-the-art models reveal substantial gaps in physical reasoning, 3D consistency, and long-horizon interaction, underscoring the need for more structured evaluation protocols for generative world models.

cs.CV↗

arXiv2026

BIFE: Better Interaction, Fewer Errors for Minute-Long Video Generation

Long video generation is a critical step toward building realistic world models, requiring both high visual fidelity and long-range interaction consistency. Recent autoregressive diffusion models enable long-horizon generation through KV cache reuse, yet suffer from two fundamental challenges: failure to preserve long-range interactions due to sliding-window KV cache and error accumulation that progressively degrades generation quality over time. To address these issues, we propose BIFE, a framework that introduces a semantic sparse KV cache for retrieval-based long-range conditioning and a Block Forcing training strategy to enforce cross-block consistency. Together, these designs preserve historical interactions while mitigating drift, enabling stable and coherent minute-long video generation. We also introduce InterVBench, a minute-long video benchmark with fine-grained block-level annotations and Video Drift Error metrics. Extensive experiments on InterVBench and VBench-Long demonstrate that BIFE achieves state-of-the-art performance, including a 22.2% improvement on VDE-Subject and a 19.4% improvement on VDE-Clarity over baselines. Website: https://alibaba-damo-academy.github.io/BIFE. Code: https://github.com/alibaba-damo-academy/BIFE.

cs.CV↗

arXiv2026

Towards Error-Free Long Video Generation

Recent advances in video generation have made minute-level synthesis possible; however, generating long videos remains challenging due to error accumulation, attribute drift, and the limited availability of long video data. In this paper, we introduce an infinite-length video generation framework that focusing on addressing these issues and produces high-quality, dynamic, and identity-consistent single-shot long videos. We first finetune a diffusion model as a video extension model on large-scale short video data to autoregressively generate temporally coherent clips. Inspired by the success of large language models (LLMs), we adopt causal attention computation between clips to further finetune this model on long video data. In this way, the tokens in one clip (short video) are computed by bidirectional attention while tokens among clips are computed by unidirectional attention. This design leverages the strengths of modern diffusion models while preserving long-term context information, effectively mitigating error accumulation and attribute drift. To achieve memory efficiency during inference, we adopt a key-value (KV) caching mechanism to maintain a constant KV memory. Furthermore, we introduce truncation-rectified flow (T-RFlow) technique to further suppress error accumulation. Experimental results demonstrate the effectiveness of our method. Our framework establishes a new benchmark for realistic and coherent minute-level video synthesis.

cs.CV↗

arXiv2026

SCOPE: Scale-Consistent One-Pass Estimation of 3D Geometry

We present SCOPE (Scale-Consistent One-Pass Estimation of 3D Geometry), a novel approach for estimating 3D geometry from extended monocular video sequences, where existing methods struggle to maintain both geometric accuracy and temporal consistency across hundreds of frames. Our approach generates affine-invariant 3D point maps with shared parameters across entire sequences, enabling consistent scale-invariant representations. We introduce three key innovations: viewpoint-invariant geometry aligning multi-perspective points in a unified reference frame; appearance-invariant learning enforcing consistency across exponential timescales; and frequency-modulated positioning enabling extrapolation to sequences vastly exceeding training length. Experiments across diverse datasets demonstrate significant improvements, reducing relative point map error by 24.2% and temporal alignment error by 34.9% on ScanNet compared to state-of-the-art methods. Our approach handles challenging scenarios with complex camera trajectories and lighting variations while efficiently processing extended sequences in a single pass. Project page: https://scope3d.github.io/.

cs.CV↗

arXiv2026

H.265/HEVC Video Steganalysis Based on CU Block Structure Gradients and IPM Mapping

Existing H.265/HEVC video steganalysis research mainly focuses on detecting the steganography based on motion vectors, intra prediction modes, and transform coefficients. However, there is currently no effective steganalysis method capable of detecting steganography based on Coding Unit (CU) block structure. To address this issue, we propose, for the first time, a H.265/HEVC video steganalysis algorithm based on CU block structure gradients and intra prediction mode mapping. The proposed method first constructs a new gradient map to explicitly describe changes in CU block structure, and combines it with a block level mapping representation of IPM. It can jointly model the structural perturbations introduced by steganography based on CU block structure. Then, we design a novel steganalysis network called GradIPMFormer, whose core innovation is an integrated architecture that combines convolutional local embedding with Transformer-based token modeling to jointly capture local CU boundary perturbations and long-range cross-CU structural dependencies, thereby effectively enhancing the capability to perceive CU block structure embedding. Experimental results show that under different quantization parameters and resolution settings, the proposed method consistently achieves superior detection performance across multiple steganography methods based on CU block structure. This study provides a new CU block structure steganalysis paradigm for H.265/HEVC and has significant research value for covert communication security detection.

eess.IV↗

arXiv2026

ClinHallu: A Benchmark for Diagnosing Stage-Wise Hallucinations in Medical MLLM Reasoning

Building trustworthy medical multimodal large language models (MLLMs) is critical for reliable clinical decision support. Existing medical hallucination benchmarks mainly focus on data collection, but often ignore where hallucinations originate within the reasoning process. We find that hallucination sources vary across samples: errors may arise from visual misrecognition, incorrect medical knowledge recall, or flawed reasoning integration. To enable source-level hallucination diagnosis, we introduce ClinHallu, a benchmark for stage-wise hallucination diagnosis in medical MLLM reasoning. ClinHallu contains 7,031 validated instances, where each instance is augmented with a structured reasoning trace decomposed into Visual Recognition, Knowledge Recall, and Reasoning Integration. We also use stage-replacement interventions to measure how correcting specific stages affects the final answer. Beyond evaluation, we show that trace-supervised fine-tuning reduces stage-wise hallucinations. ClinHallu provides a fine-grained hallucination testbed for diagnosing and mitigating reasoning failures in medical MLLMs. The benchmark is publicly available at https://github.com/alibaba-damo-academy/ClinHallu.

cs.CV↗

arXiv2026

K-Forcing: Joint Next-K-Token Decoding via Push-Forward Language Modeling

Autoregressive (AR) language modeling is the dominant paradigm for text generation, yet its sequential token-by-token decoding makes inference memory-bound and inefficient. Existing acceleration approaches, such as speculative decoding and diffusion language models, can yield speedups under certain conditions but do not directly address high-load batch serving--the scenario most critical for industrial-scale deployment. We introduce K-Forcing, a push-forward language modeling paradigm for joint next-k-token decoding. K-Forcing distills an existing AR model into a conditional push-forward mapping--one that transforms independent uniform noise variables into a joint sample of multiple future tokens in a single forward pass. This design preserves fixed-length outputs, reuses the AR teacher backbone, and remains compatible with standard AR serving infrastructure. We train this mapping via progressive self-forcing distillation, which gradually expands the prediction window while enabling the student to closely match the sequence distribution of the AR teacher. We evaluate K-Forcing on LM1B and OpenWebText using a standard causal Transformer backbone. When aggressively configured to generate k = 4 tokens per forward pass, K-Forcing delivers approximately 2.4-3.5x speedup across different batch sizes, while incurring modest quality degradation relative to its AR teacher. As inference increasingly dominates the lifetime compute cost of modern LLMs, K-Forcing offers a promising route toward accelerating AR generation under real-world high-load deployment.

cs.LG↗

arXiv2026

A Kernel-Clean Lean Mechanization of Classical Lottery in Action and the Wakker--Debreu--Koopmans Representation Layer

We present a Lean 4/Mathlib formalization of the additive representation theory behind Classical Lottery in Action and the Wakker-Debreu-Koopmans (WDK) layer it relies on. Our central result is a machine-checked proof that the cross-pair Thomsen / double-cancellation (hexagon) condition is irreducible from the ordinal axioms of additive conjoint measurement (weak order, restricted solvability, Archimedean condition, and tradeoff consistency). We exhibit an explicit verified counter-model (additiveRealBoolPref) satisfying all ordinal axioms yet failing the cross-pair condition, with every strict standard sequence being an arithmetic progression and hence non-dense. Around this boundary we mechanize the full derivable construction: continuous Debreu/Eilenberg utility from separability, standard-sequence grids, bisection methods from connectedness, and global additive gluing. All public theorems are sorry-free conditional wrappers over this single irreducible structural input. The development is kernel-clean, depending only on standard Lean foundations (propext, Classical.choice, Quot.sound). The companion file ClassicalLotteryInAction.lean formalizes local classical-lottery constructions, average-utility results, matching-frequency lemmas, and ambiguity-attitude statements used by the Management Science paper. This draws a precise, machine-certified line between what additive conjoint measurement can prove and what it must assume.

cs.LO↗

arXiv2026

Towards 3D-Aware Video Diffusion Models: Render-Free Human Motion Control with Mesh Tokenization

Diffusion models have shown remarkable success in video generation. However, whether such models are truly aware of the 3D structure underlying visual observations, rather than simply reproducing plausible 2D projections, remains an open question. In this work, we investigate this question through human motion control, a task that requires precise modelling of 3D human geometry, motion, camera viewpoint, and scene context. Unlike prior methods that rely on rendered 2D motion guidance videos, we propose a render-free framework that conditions video generation directly on compressed 3D human mesh tokens. This representation preserves full 3D geometric information while enabling a unified token-based generation pipeline that processes video tokens jointly with motion tokens in a DiT-based architecture. This design requires the model to reason jointly about appearance, 3D structure, and camera viewpoint during video generation. Experimental results demonstrate strong performance on human motion control benchmarks, while reducing artifacts induced by view-dependent 2D guidance and trajectory-pose mismatches during editing. These findings suggest that video diffusion models, when equipped with mesh tokenization, can better capture complex 3D human structures and their interactions with the surrounding environment.

cs.CV↗

arXiv2026

RynnVLA-002: A Unified Vision-Language-Action and World Model

We introduce RynnVLA-002, a unified Vision-Language-Action (VLA) and world model. The world model leverages action and visual inputs to predict future image states, learning the underlying physics of the environment to refine action generation. Conversely, the VLA model produces subsequent actions from image observations, enhancing visual understanding and supporting the world model's image generation. The unified framework of RynnVLA-002 enables joint learning of environmental dynamics and action planning. Our experiments show that RynnVLA-002 surpasses individual VLA and world models, demonstrating their mutual enhancement. We evaluate RynnVLA-002 in both simulation and real-world robot tasks. RynnVLA-002 achieves 97.4% success rate on the LIBERO simulation benchmark without pretraining, while in real-world LeRobot experiments, its integrated world model boosts the overall success rate by 50%.

cs.RO↗

arXiv2026

DGS-Net: Distillation-Guided Gradient Surgery for CLIP Fine-Tuning in AI-Generated Image Detection

The rapid progress of generative models such as GANs and diffusion models has led to the widespread proliferation of AI-generated images, raising concerns about misinformation, privacy violations, and trust erosion in digital media. Although large-scale multimodal models like CLIP offer strong transferable representations for detecting synthetic content, fine-tuning them often induces catastrophic forgetting, which degrades pre-trained priors and limits cross-domain generalization. To address this issue, we propose the Distillation-guided Gradient Surgery Network (DGS-Net), a novel framework that preserves transferable pre-trained priors while suppressing task-irrelevant components. Specifically, we introduce a gradient-space decomposition that separates harmful and beneficial descent directions during optimization. By projecting task gradients onto the orthogonal complement of harmful directions and aligning with beneficial ones distilled from a frozen CLIP encoder, DGS-Net achieves unified optimization of prior preservation and irrelevant suppression. Extensive experiments on 50 generative models demonstrate that our method outperforms state-of-the-art approaches by an average margin of 6.6, achieving superior detection performance and generalization across diverse generation techniques.

cs.CV↗

arXiv2026

MieDB-100k: A Comprehensive Dataset for Medical Image Editing

The scarcity of high-quality data remains a primary bottleneck in adapting multimodal generative models for medical image editing. Existing medical image editing datasets often suffer from limited diversity, neglect of medical image understanding and inability to balance quality with scalability. To address these gaps, we propose MieDB-100k, a large-scale, high-quality and diverse dataset for text-guided medical image editing. It categorizes editing tasks into perspectives of Perception, Modification and Transformation, considering both understanding and generation abilities. We construct MieDB-100k via a data curation pipeline leveraging both modality-specific expert models and rule-based data synthetic methods, followed by rigorous manual inspection to ensure clinical fidelity. Extensive experiments demonstrate that model trained with MieDB-100k consistently outperform both open-source and proprietary models while exhibiting strong generalization ability. We anticipate that this dataset will serve as a cornerstone for future advancements in specialized medical image editing.

cs.CV↗

arXiv2026

OpenAI o1 System Card

The o1 model series is trained with large-scale reinforcement learning to reason using chain of thought. These advanced reasoning capabilities provide new avenues for improving the safety and robustness of our models. In particular, our models can reason about our safety policies in context when responding to potentially unsafe prompts, through deliberative alignment. This leads to state-of-the-art performance on certain benchmarks for risks such as generating illicit advice, choosing stereotyped responses, and succumbing to known jailbreaks. Training models to incorporate a chain of thought before answering has the potential to unlock substantial benefits, while also increasing potential risks that stem from heightened intelligence. Our results underscore the need for building robust alignment methods, extensively stress-testing their efficacy, and maintaining meticulous risk management protocols. This report outlines the safety work carried out for the OpenAI o1 and OpenAI o1-mini models, including safety evaluations, external red teaming, and Preparedness Framework evaluations.

cs.AI↗

arXiv2026

Online Learning for Autoregressive Multilayer Stochastic Block Models under Stationarity and Non-Stationarity

Dynamic multilayer networks arise in many applications where multiple types of relations among a common set of nodes evolve over time. Existing approaches often assume temporal independence, focus on single-layer networks or impose stationarity, limiting their applicability in practice. In this paper, we introduce a first-order autoregressive multilayer stochastic block model (AR(1)-MSBM), in which edge formation and dissolution probabilities between consecutive time points are determined by latent community memberships and shared across layers. Under stationarity, we propose an online estimation procedure based on recursive updates and tensor-based spectral refinement. We establish non-asymptotic estimation rates, prove their minimax optimality and derive guarantees for community recovery. We further consider a non-stationary setting that allows both abrupt changes and gradual shifts, and develop an adaptive windowed online algorithm that automatically adjusts to unknown structural changes. Under a quasi-stationary segmentation framework, we derive estimation and community recovery guarantees that match the stationary results when applied segmentwise. Our theoretical findings are supported by extensive numerical experiments, with code available online.

stat.ME↗

arXiv2026

Empirical Insights of Test Selection Metrics under Multiple Testing Objectives and Distribution Shifts

Deep learning (DL)-based systems can exhibit unexpected behavior when exposed to out-of-distribution (OOD) scenarios, posing serious risks in safety-critical domains such as malware detection and autonomous driving. This underscores the importance of thoroughly testing such systems before deployment. To this end, researchers have proposed a wide range of test selection metrics designed to effectively select inputs. However, prior evaluations of metrics reveal three key limitations: (1) narrow testing objectives, for example, many studies assess metrics only for fault detection, leaving their effectiveness for performance estimation unclear; (2) limited coverage of OOD scenarios, with natural and label shifts are rarely considered; (3) Biased dataset selection, where most work focuses on image data while other modalities remain underexplored. Consequently, a unified benchmark that examines how these metrics perform under multiple testing objectives, diverse OOD scenarios, and different data modalities is still lacking. This leaves practitioners uncertain about which test selection metrics are most suitable for their specific objectives and contexts. To address this gap, we conduct an extensive empirical study of 15 existing metrics, evaluating them under three testing objectives (fault detection, performance estimation, and retraining guidance), five types of OOD scenarios (corrupted, adversarial, temporal, natural, and label shifts), three data modalities (image, text, and Android packages), and 13 DL models. In total, our study encompasses 1,640 experimental scenarios, offering a comprehensive evaluation and statistical analysis.

cs.SE↗

arXiv2026

CamPVG: Camera-Controlled Panoramic Video Generation with Epipolar-Aware Diffusion

Recently, camera-controlled video generation has seen rapid development, offering more precise control over video generation. However, existing methods predominantly focus on camera control in perspective projection video generation, while geometrically consistent panoramic video generation remains challenging. This limitation is primarily due to the inherent complexities in panoramic pose representation and spherical projection. To address this issue, we propose CamPVG, the first diffusion-based framework for panoramic video generation guided by precise camera poses. We achieve camera position encoding for panoramic images and cross-view feature aggregation based on spherical projection. Specifically, we propose a panoramic Plücker embedding that encodes camera extrinsic parameters through spherical coordinate transformation. This pose encoder effectively captures panoramic geometry, overcoming the limitations of traditional methods when applied to equirectangular projections. Additionally, we introduce a spherical epipolar module that enforces geometric constraints through adaptive attention masking along epipolar lines. This module enables fine-grained cross-view feature aggregation, substantially enhancing the quality and consistency of generated panoramic videos. Extensive experiments demonstrate that our method generates high-quality panoramic videos consistent with camera trajectories, far surpassing existing methods in panoramic video generation.

cs.CV↗

arXiv2026

How Noise Benefits AI-generated Image Detection

The rapid advancement of generative models has made real and synthetic images increasingly indistinguishable. Although extensive efforts have been devoted to detecting AI-generated images, out-of-distribution generalization remains a persistent challenge. We trace this weakness to spurious shortcuts exploited during training and we also observe that small feature-space perturbations can mitigate shortcut dominance. To address this problem in a more controllable manner, we propose the Positive-Incentive Noise for CLIP (PiN-CLIP), which jointly trains a noise generator and a detection network under a variational positive-incentive principle. Specifically, we construct positive-incentive noise in the feature space via cross-attention fusion of visual and categorical semantic features. During optimization, the noise is injected into the feature space to fine-tune the visual encoder, suppressing shortcut-sensitive directions while amplifying stable forensic cues, thereby enabling the extraction of more robust and generalized artifact representations. Comparative experiments are conducted on an open-world dataset comprising synthetic images generated by 42 distinct generative models. Our method achieves new state-of-the-art performance, with notable improvements of 5.4 in average accuracy over existing approaches.

cs.CV↗

arXiv2026

The Stellar "Snake"-III: Co-evolution of Stars and Molecular Clouds Unveiled by Gaia, MWISP, and LAMOST

By combining multi-band data from Gaia DR3, MWISP CO, and LAMOST DR11 LSR/MSR, we investigate the co-evolution of stars and their parent molecular cloud in a snake-like stellar structure, named Snake III. Based on 5-D phase-space selection, we identified 5683 member stars (median age 7.6 Myr) across approximately $300 \times 500 \times 175$ pc$^3$ volume, along with 12 embedded open clusters. Then we use BEEP distances combined with $^{12}$CO velocities to clearly identify the molecular clouds associated with the stellar complex in spatial and kinematics. The molecular cloud density increases with Galactic longitude, with older open clusters forming in cavities near higher-density regions (except ASCC 125), while young field stars currently form preferentially in present-day high-density environments, indicating that cloud density regulates the star-formation sequence. $^{12}$CO excitation temperature, centroid velocity, velocity dispersion and H$α$ emission reveal that early feedback first compresses cloud edges to trigger new stars, then sweeps and disperses the parent clouds. The extremely young cluster (ASCC 125, 4.4 Myr) lies near the densest region yet is surrounded by a shell with bidirectional density-velocity perturbations, consistent with a delayed-triggering scenario under the combined influence of UBC 178 stellar-wind feedback and a suspected supernova blast. Our results naturally demonstrate that snake-like stellar structures are filamentary relics of hierarchical star formation within giant molecular clouds. They provide direct observational evidence that cloud density and early feedback jointly modulate the progression of star formation, offering a clear and young laboratory for studying star-cloud co-evolution.

astro-ph.GA↗