Changsen Yuan

Haitian Li, Yanghao Zhou, Heyan Huang et al.

In recent years, Multi-Talker Audio-Video Generation (MTAVG) models have shown promising performance on fundamental metrics such as lip-sync and audio-visual alignment. However, these metrics remain insufficient for assessing cinematic expressiveness in scene-level generation. In multi-character scenes, generation models must go beyond audio-visual realism to convey coherent character performance and other higher-level cinematic qualities. To fill this gap, we introduce MTAVG-Bench 2.0, a benchmark for diagnosing failure modes of cinematic expressiveness in multi-talker audio-video generation. Unlike prior settings that mainly focus on the quality of basic multi-turn dialogue, MTAVG-Bench 2.0 targets short-drama and scene-level generation, and establishes a high-level failure taxonomy spanning acting, narrative, atmosphere, and audio-visual language. Based on this taxonomy, we construct more than 10,000 question-answering evaluation instances, together with subsets for short-drama-level assessment and temporal localization of failure modes, to systematically evaluate the ability of omni large language models to diagnose high-level audio-visual failures. Experimental results show that commercial omni models such as Gemini substantially outperform other evaluators, yet even the strongest models continue to struggle with complex failures in our benchmark. These results demonstrate that MTAVG-Bench 2.0 provides a systematic benchmark for failure diagnosis in cinematic multi-talker audio-video generation.

Yang-Hao Zhou, Haitian Li, Rexar Lin et al.

Recent advances in text-to-audio-video (T2AV) generation have enabled models to synthesize audio-visual videos with multi-participant dialogues. However, existing evaluation benchmarks remain largely designed for human-recorded videos or single-speaker settings. As a result, structural failures in generated multi-talker dialogue videos, such as identity drift, unnatural turn transitions, and audio-visual misalignment, cannot be effectively diagnosed. To address this issue, we introduce MTAVG-Bench, a failure-driven diagnostic benchmark for multi-talker dialogue-centric audio-video generation. MTAVG-Bench is built via a semi-automatic pipeline, where 1.8k videos are generated using mainstream T2AV models with carefully designed prompts, yielding 2.4k manually annotated QA pairs for fine-grained failure diagnosis. The benchmark evaluates multi-speaker dialogue generation at four levels: audio-visual signal fidelity, temporal attribute consistency, social interaction, and cinematic expression. Built on a hierarchical failure taxonomy and a targeted QA protocol, MTAVG-Bench is primarily designed to evaluate whether proprietary and open-source omni-models can reliably identify failure modes in multi-speaker T2AV outputs. We benchmark 12 proprietary and open-source omni-models on MTAVG-Bench, with Gemini 3 Pro achieving the strongest overall performance, while leading open-source models remain competitive in signal fidelity and consistency. Overall, MTAVG-Bench enables fine-grained failure analysis for rigorous model comparison and targeted video generation refinement.

Xiao Liu, Heyan Huang, Yue Zhang et al.

We consider direct modeling of underlying stock value movement sequences over time in the news-driven stock movement prediction. A recurrent state transition model is constructed, which better captures a gradual process of stock movement continuously by modeling the correlation between past and future price movements. By separating the effects of news and noise, a noisy random factor is also explicitly fitted based on the recurrent states. Results show that the proposed model outperforms strong baselines. Thanks to the use of attention over news events, our model is also more explainable. To our knowledge, we are the first to explicitly model both events and noise over a fundamental stock value state for news-driven stock movement prediction.

Changsen Yuan, Heyan Huang, Chong Feng et al.

Distant supervision for relation extraction is an efficient method to reduce labor costs and has been widely used to seek novel relational facts in large corpora, which can be identified as a multi-instance multi-label problem. However, existing distant supervision methods suffer from selecting important words in the sentence and extracting valid sentences in the bag. Towards this end, we propose a novel approach to address these problems in this paper. Firstly, we propose a linear attenuation simulation to reflect the importance of words in the sentence with respect to the distances between entities and words. Secondly, we propose a non-independent and identically distributed (non-IID) relevance embedding to capture the relevance of sentences in the bag. Our method can not only capture complex information of words about hidden relations, but also express the mutual information of instances in the bag. Extensive experiments on a benchmark dataset have well-validated the effectiveness of the proposed method.