arXiv subjects

Liwei Che

Publications and source records attributed to Liwei Che.

12 recordsLinked to original sources

arXiv2026

SignVerse-2M: A Two-Million-Clip Pose-Native Universe of 55+ Sign Languages

Large-scale sign language resources usually pair RGB videos with text and support recognition, translation, and generation. However, RGB frames contain backgrounds, clothing, and other appearance information unrelated to signing motion. Representing the videos as whole-body keypoints such as DWPose removes much of this variation, produces a compact description of body, hand, and facial movement, and matches the input used by recent pose-guided video models. Existing datasets do not provide this representation at scale for in-the-wild sign language videos. We present SignVerse-2M, a multilingual dataset containing approximately two million pose clips from more than 55 sign languages. We process publicly available sign language videos with a shared DWPose pipeline, representing body, hand, and facial motion in the same keypoint format across all clips. The source videos include diverse signers, viewpoints, and recording environments. We also release the processing pipeline, task definitions, and SignDW Transformer baselines. Experiments show that the data supports multilingual text-to-pose modeling and that the generated poses can directly drive several recent pose-conditioned rendering systems.

cs.CV↗

arXiv2026

Stable Signer V3: 4D Deformable Primitive Splatting Is All You Need for Sign Language Production

We present Stable Signer V3, an interactive high-resolution sign language avatar system built on 4D Deformable Primitive Splatting. It addresses three limitations of existing avatars: restricted facial expression, inaccurate identity-specific body deformation, and loss of fine appearance details. Shared Face Primitive Splatting adds transferable superquadric primitives for richer expression control; Test-Time Body Primitive Splatting learns an identity-specific virtual muscle layer for more faithful motion; and Test-Time Tunneling Super-Resolution uses TunnelingNet to recover identity-specific high-resolution details. Both test-time modules are optimized once per signer identity; inference then runs entirely from the stored identity parameters. Together, they reduce facial geometry error by 71.4%, body-boundary error by 35.1% across ten identities, and high-resolution RGB reconstruction error by 24.2% on held-out evaluations, while retaining interactive streaming.

cs.CV↗

arXiv2026

Seeing, Maintaining, and Learning: The Evolution of Multimodal Agent Memory

Multimodal models and agents have emerged to enable AI systems to learn from and interact with the real world more like humans by integrating information across multiple modalities. However, enabling multimodal agents to continually acquire useful information from complex, heterogeneous multimodal data remains a significant challenge as they move toward long-horizon tasks and lifelong settings. Addressing this challenge requires effective memory mechanisms that can maintain continuity across sessions, adapt to changing users and environments, and refine memory-management policies through accumulated experience. We introduce a two-axis analytical framework for multimodal agent memory structured around two key questions: how multimodal information is formed and represented (Axis I, Formation and Representation), and how its management adapts with accumulated experience (Axis II, Memory Management). Axis I, Formation and Representation, characterizes what information is retained and how it is encoded (e.g., modality-specific embeddings, textual and structured records, latent and parametric representations) to support subsequent reasoning and action. Axis II, Memory Management, distinguishes three tiers: T1 Static Memory reuses fixed memory; T2 Dynamic Memory updates memory through a stable policy; and T3 Self-Evolving Memory updates memory through an evolving policy. Based on this framework, we synthesize recurring design patterns and derive principles for balancing source fidelity with abstraction, maintaining temporal validity and provenance, grounding memory use in recoverable evidence, and evolving memory policies over time. Furthermore, we discuss the challenges and potential future directions for the memory mechanism of multimodal agents.

cs.AI↗

arXiv2026

Why MLLMs Struggle to Count: Overcoming Individuation and Aggregation Bottlenecks with ConvStack

Multimodal Large Language Models (MLLMs) consistently struggle with fine-grained visual counting, yet the underlying causes remain poorly understood. In this work, we present a mechanistic analysis of this failure mode, identifying two critical bottlenecks inherent to the global attention pipeline of MLLMs. First, we reveal an individuation bottleneck stemming from image patchification: because Vision Transformers process patches independently, they struggle to group fragmented geometric features across boundaries into distinct object representations. Second, we identify a collapse in the subsequent counting aggregation process, where representation separation rapidly diminishes as numerosity increases due to attention compression. Identifying and formalizing these twin bottlenecks constitutes our first major contribution. To overcome them, we propose ConvStack, a lightweight architecture that operates directly in the visual token space to explicitly aggregate and inject local spatial structures via zero-initialized residual connections. By explicitly addressing the individuation bottleneck, ConvStack provides unambiguous geometric evidence for downstream aggregation. Remarkably, by fine-tuning exclusively on counting tasks, the model achieves substantial improvements in dense object counting and broader spatial understanding benchmarks, without compromising on general visual capabilities.

cs.CV↗

arXiv2026

LVMH: A Meta-Optimization Harness for Long-Video Agent Memory

Long-video agents must map hours or days of multimodal evidence into a bounded context while respecting token and latency budgets, making memory design a context-management challenge. Existing memory systems largely rely on hand-designed policies, leaving open how to jointly optimize memory design for performance and efficiency. We propose Long Video Meta-memory Harness (LVMH), an automated multi-agent framework that formulates memory-policy development as multi-objective code search around a fixed answer model. To guard against self-certification and development-set overfitting, LVMH separates implementation from independent review and enforces protected observation boundaries. Persistent meta-memory records past proposals, traces, costs, and failures to guide subsequent policy revisions. Through this harness, we discover Self-Evolving Evidence Memory (SEEM), a "text-first, vision-second" policy that answers from compact structured event memory and evidence specialists, consulting localized frames through a selective visual gate. The policy evolved on EgoLifeQA generalizes to unseen benchmarks, outperforming Meta-Harness on both Ego-R1 Bench and Video-MME (L) with GPT-5 mini. Across the three benchmarks, SEEM with GPT-5 achieves 65.7% average accuracy, surpassing the previous state of the art by 7.7 percentage points with approximately 32x lower average evaluation token use and 91x shorter average response time.

cs.CV↗

arXiv2026

MemEye: A Visual-Centric Evaluation Framework for Multimodal Agent Memory

Long-term agent memory is increasingly multimodal, yet existing evaluations rarely test whether agents preserve the visual evidence needed for later reasoning. In prior work, many visually grounded questions can be answered using only captions or textual traces, allowing answers to be inferred without preserving the fine-grained visual evidence. Meanwhile, harder cases that require reasoning over changing visual states are largely absent. Therefore, we introduce MemEye, a framework that evaluates memory capabilities from two dimensions: one measures the granularity of decisive visual evidence (from scene-level to pixel-level evidence), and the other measures how retrieved evidence must be used (from single evidence to evolutionary synthesis). Under this framework, we construct a new benchmark across 8 life-scenario tasks, with ablation-driven validation gates for assessing answerability, shortcut resistance, visual necessity, and reasoning structure. By evaluating 13 memory methods across 4 VLM backbones, we show that current architectures still struggle to preserve fine-grained visual details and reason about state changes over time. Our findings show that long-term multimodal memory depends on evidence routing, temporal tracking, and detail extraction.

cs.CV↗

arXiv2026

Counting Circuits: Mechanistic Interpretability of Visual Reasoning in Large Vision-Language Models

Counting serves as a simple but powerful test of a Large Vision-Language Model's (LVLM's) reasoning; it forces the model to identify each individual object and then add them all up. In this study, we investigate how LVLMs implement counting using controlled synthetic and real-world benchmarks, combined with mechanistic analyses. Our results show that LVLMs display a human-like counting behavior, with precise performance on small numerosities and noisy estimation for larger quantities. We introduce two novel interpretability methods, Visual Activation Patching and HeadLens, and use them to uncover a structured "counting circuit" that is largely shared across a variety of visual reasoning tasks. Building on these insights, we propose a lightweight intervention strategy that exploits simple and abundantly available synthetic images to fine-tune arbitrary pretrained LVLMs exclusively on counting. Despite the narrow scope of this fine-tuning, the intervention not only enhances counting accuracy on in-distribution synthetic data, but also yields an average improvement of +8.36% on out-of-distribution counting benchmarks and an average gain of +1.54% on complex, general visual reasoning tasks for Qwen2.5-VL. These findings highlight the central, influential role of counting in visual reasoning and suggest a potential pathway for improving overall visual reasoning capabilities through targeted enhancement of counting mechanisms.

cs.CV↗

arXiv2025

Hallucinatory Image Tokens: A Training-free EAZY Approach on Detecting and Mitigating Object Hallucinations in LVLMs

Despite their remarkable potential, Large Vision-Language Models (LVLMs) still face challenges with object hallucination, a problem where their generated outputs mistakenly incorporate objects that do not actually exist. Although most works focus on addressing this issue within the language-model backbone, our work shifts the focus to the image input source, investigating how specific image tokens contribute to hallucinations. Our analysis reveals a striking finding: a small subset of image tokens with high attention scores are the primary drivers of object hallucination. By removing these hallucinatory image tokens (only 1.5% of all image tokens), the issue can be effectively mitigated. This finding holds consistently across different models and datasets. Building on this insight, we introduce EAZY, a novel, training-free method that automatically identifies and Eliminates hAllucinations by Zeroing out hallucinatorY image tokens. We utilize EAZY for unsupervised object hallucination detection, achieving 15% improvement compared to previous methods. Additionally, EAZY demonstrates remarkable effectiveness in mitigating hallucinations while preserving model utility and seamlessly adapting to various LVLM architectures.

cs.CV↗

arXiv2024

Leveraging Foundation Models for Multi-modal Federated Learning with Incomplete Modality

Federated learning (FL) has obtained tremendous progress in providing collaborative training solutions for distributed data silos with privacy guarantees. However, few existing works explore a more realistic scenario where the clients hold multiple data modalities. In this paper, we aim to solve a novel challenge in multi-modal federated learning (MFL) -- modality missing -- the clients may lose part of the modalities in their local data sets. To tackle the problems, we propose a novel multi-modal federated learning method, Federated Multi-modal contrastiVe training with Pre-trained completion (FedMVP), which integrates the large-scale pre-trained models to enhance the federated training. In the proposed FedMVP framework, each client deploys a large-scale pre-trained model with frozen parameters for modality completion and representation knowledge transfer, enabling efficient and robust local training. On the server side, we utilize generated data to uniformly measure the representation similarity among the uploaded client models and construct a graph perspective to aggregate them according to their importance in the system. We demonstrate that the model achieves superior performance over two real-world image-text classification datasets and is robust to the performance degradation caused by missing modality.

cs.LG↗

arXiv2023

Towards Personalized Federated Learning via Heterogeneous Model Reassembly

This paper focuses on addressing the practical yet challenging problem of model heterogeneity in federated learning, where clients possess models with different network structures. To track this problem, we propose a novel framework called pFedHR, which leverages heterogeneous model reassembly to achieve personalized federated learning. In particular, we approach the problem of heterogeneous model personalization as a model-matching optimization task on the server side. Moreover, pFedHR automatically and dynamically generates informative and diverse personalized candidates with minimal human intervention. Furthermore, our proposed heterogeneous model reassembly technique mitigates the adverse impact introduced by using public data with different distributions from the client data to a certain extent. Experimental results demonstrate that pFedHR outperforms baselines on three datasets under both IID and Non-IID settings. Additionally, pFedHR effectively reduces the adverse impact of using different public data and dynamically generates diverse personalized models in an automated manner.

cs.LG↗

arXiv2021

FedTriNet: A Pseudo Labeling Method with Three Players for Federated Semi-supervised Learning

Federated Learning has shown great potentials for the distributed data utilization and privacy protection. Most existing federated learning approaches focus on the supervised setting, which means all the data stored in each client has labels. However, in real-world applications, the client data are impossible to be fully labeled. Thus, how to exploit the unlabeled data should be a new challenge for federated learning. Although a few studies are attempting to overcome this challenge, they may suffer from information leakage or misleading information usage problems. To tackle these issues, in this paper, we propose a novel federated semi-supervised learning method named FedTriNet, which consists of two learning phases. In the first phase, we pre-train FedTriNet using labeled data with FedAvg. In the second phase, we aim to make most of the unlabeled data to help model learning. In particular, we propose to use three networks and a dynamic quality control mechanism to generate high-quality pseudo labels for unlabeled data, which are added to the training set. Finally, FedTriNet uses the new training set to retrain the model. Experimental results on three publicly available datasets show that the proposed FedTriNet outperforms state-of-the-art baselines under both IID and Non-IID settings.

cs.LG↗

arXiv2021

FedSiam: Towards Adaptive Federated Semi-Supervised Learning

Federated learning (FL) has emerged as an effective technique to co-training machine learning models without actually sharing data and leaking privacy. However, most existing FL methods focus on the supervised setting and ignore the utilization of unlabeled data. Although there are a few existing studies trying to incorporate unlabeled data into FL, they all fail to maintain performance guarantees or generalization ability in various real-world settings. In this paper, we focus on designing a general framework FedSiam to tackle different scenarios of federated semi-supervised learning, including four settings in the labels-at-client scenario and two setting in the labels-at-server scenario. FedSiam is built upon a siamese network into FL with a momentum update to handle the non-IID challenges introduced by unlabeled data. We further propose a new metric to measure the divergence of local model layers within the siamese network. Based on the divergence, FedSiam can automatically select layer-level parameters to be uploaded to the server in an adaptive manner. Experimental results on three datasets under two scenarios with different data distribution settings demonstrate that the proposed FedSiam framework outperforms state-of-the-art baselines.

cs.LG↗