Liang Yi

Zhongjie Ba, Liang Yi, Peng Cheng et al.

Toxic speech detection has become a crucial challenge in maintaining safe online communication environments. However, existing approaches to toxic speech detection often neglect the contribution of paralinguistic cues, such as emotion, intonation, and speech rate, which are key to detecting speech toxicity. Moreover, current toxic speech datasets are predominantly text-based, limiting the development of models that can capture paralinguistic cues.To address these challenges, we present ToxiAlert-Bench, a large-scale audio dataset comprising over 30,000 audio clips annotated with seven major toxic categories and twenty fine-grained toxic labels. Uniquely, our dataset annotates toxicity sources -- distinguishing between textual content and paralinguistic origins -- for comprehensive toxic speech analysis.Furthermore, we propose a dual-head neural network with a multi-stage training strategy tailored for toxic speech detection. This architecture features two task-specific classification headers: one for identifying the source of sensitivity (textual or paralinguistic), and the other for categorizing the specific toxic type. The training process involves independent head training followed by joint fine-tuning to reduce task interference. To mitigate data class imbalance, we incorporate class-balanced sampling and weighted loss functions.Our experimental results show that leveraging paralinguistic features significantly improves detection performance. Our method consistently outperforms existing baselines across multiple evaluation metrics, with a 21.1% relative improvement in Macro-F1 score and a 13.0% relative gain in accuracy over the strongest baseline, highlighting its enhanced effectiveness and practical applicability.

Xingyu Zhu, Liang Yi, Shuo Wang et al.

Multimodal 3D vision-language models show strong generalization across diverse 3D tasks, but their performance still degrades notably under domain shifts. This has motivated recent studies on test-time adaptation (TTA), which enables models to adapt online using test-time data. Among existing TTA methods, cache-based mechanisms are widely adopted for leveraging previously observed samples in online prediction refinement. However, they store only limited historical information, leading to progressive information loss as the test stream evolves. In addition, their prediction logits are fused heuristically, making adaptation unstable. To address these limitations, we propose BayesMM, a Multimodal Bayesian Distribution Learning framework for test-time point cloud analysis. BayesMM models textual priors and streaming visual features of each class as Gaussian distributions: textual parameters are derived from semantic prompts, while visual parameters are updated online with arriving samples. The two modalities are fused via Bayesian model averaging, which automatically adjusts their contributions based on posterior evidence, yielding a unified prediction that adapts continually to evolving test-time data without training. Extensive experiments on multiple point cloud benchmarks demonstrate that BayesMM maintains robustness under distributional shifts, yielding over 4% average improvement.

Junpei Liao, Zhikai Chen, Liang Yi et al.

Deep learning based video frame interpolation (VIF) method, aiming to synthesis the intermediate frames to enhance video quality, have been highly developed in the past few years. This paper investigates the adversarial robustness of VIF models. We apply adversarial attacks to VIF models and find that the VIF models are very vulnerable to adversarial examples. To improve attack efficiency, we suggest to make full use of the property of video frame interpolation task. The intuition is that the gap between adjacent frames would be small, leading to the corresponding adversarial perturbations being similar as well. Then we propose a novel attack method named Inter-frame Accelerate Attack (IAA) that initializes the perturbation as the perturbation for the previous adjacent frame and reduces the number of attack iterations. It is shown that our method can improve attack efficiency greatly while achieving comparable attack performance with traditional methods. Besides, we also extend our method to video recognition models which are higher level vision tasks and achieves great attack efficiency.