Yanghao Zhou

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.

Sen Fang, Chunyu Sui, Yanghao Zhou et al.

In this paper, we propose a dual-condition diffusion pre-training model named SignDiff that can generate human sign language speakers from a skeleton pose. SignDiff has a novel Frame Reinforcement Network called FR-Net, similar to dense human pose estimation work, which enhances the correspondence between text lexical symbols and sign language dense pose frames, reduces the occurrence of multiple fingers in the diffusion model. In addition, we propose a new method for American Sign Language Production (ASLP), which can generate ASL skeletal pose videos from text input, integrating two new improved modules and a new loss function to improve the accuracy and quality of sign language skeletal posture and enhance the ability of the model to train on large-scale data. We propose the first baseline for ASL production and report the scores of 17.19 and 12.85 on BLEU-4 on the How2Sign dev/test sets. We evaluated our model on the previous mainstream dataset PHOENIX14T, and the experiments achieved the SOTA results. In addition, our image quality far exceeds all previous results by 10 percentage points in terms of SSIM.

Cailing Han, Zhangbin Li, Jinxing Zhou et al.

Point-level weakly-supervised temporal sentiment localization (P-WTSL) aims to detect sentiment-relevant segments in untrimmed multimodal videos using timestamp sentiment annotations, which greatly reduces the costly frame-level labeling. To further tackle the challenges of imprecise sentiment boundaries in P-WTSL, we propose the Face-guided Sentiment Boundary Enhancement Network (\textbf{FSENet}), a unified framework that leverages fine-grained facial features to guide sentiment localization. Specifically, our approach \textit{first} introduces the Face-guided Sentiment Discovery (FSD) module, which integrates facial features into multimodal interaction via dual-branch modeling for effective sentiment stimuli clues; We \textit{then} propose the Point-aware Sentiment Semantics Contrast (PSSC) strategy to discriminate sentiment semantics of candidate points (frame-level) near annotation points via contrastive learning, thereby enhancing the model's ability to recognize sentiment boundaries. At \textit{last}, we design the Boundary-aware Sentiment Pseudo-label Generation (BSPG) approach to convert sparse point annotations into temporally smooth supervisory pseudo-labels. Extensive experiments and visualizations on the benchmark demonstrate the effectiveness of our framework, achieving state-of-the-art performance under full supervision, video-level, and point-level weak supervision, thereby showcasing the strong generalization ability of our FSENet across different annotation settings.

MiniCPM Team, Chaojun Xiao, Yuxuan Li et al. · tencent-ai, tsinghua

This paper introduces MiniCPM4, a highly efficient large language model (LLM) designed explicitly for end-side devices. We achieve this efficiency through systematic innovation in four key dimensions: model architecture, training data, training algorithms, and inference systems. Specifically, in terms of model architecture, we propose InfLLM v2, a trainable sparse attention mechanism that accelerates both prefilling and decoding phases for long-context processing. Regarding training data, we propose UltraClean, an efficient and accurate pre-training data filtering and generation strategy, and UltraChat v2, a comprehensive supervised fine-tuning dataset. These datasets enable satisfactory model performance to be achieved using just 8 trillion training tokens. Regarding training algorithms, we propose ModelTunnel v2 for efficient pre-training strategy search, and improve existing post-training methods by introducing chunk-wise rollout for load-balanced reinforcement learning and data-efficient tenary LLM, BitCPM. Regarding inference systems, we propose CPM.cu that integrates sparse attention, model quantization, and speculative sampling to achieve efficient prefilling and decoding. To meet diverse on-device requirements, MiniCPM4 is available in two versions, with 0.5B and 8B parameters, respectively. Furthermore, we construct a hybrid reasoning model, MiniCPM4.1, which can be used in both deep reasoning mode and non-reasoning mode. Evaluation results demonstrate that MiniCPM4 and MiniCPM4.1 outperform similar-sized open-source models across benchmarks, with the 8B variants showing significant speed improvements on long sequence understanding and generation.

Jinxing Zhou, Yanghao Zhou, Yaoting Wang et al.

Language-referred audio-visual segmentation (Ref-AVS) aims to segment target objects described by natural language by jointly reasoning over video, audio, and text. Beyond generating segmentation masks, providing rich and interpretable diagnoses of mask quality remains largely underexplored. In this work, we introduce Mask Quality Assessment in the Ref-AVS context (MQA-RefAVS), a new task that evaluates the quality of candidate segmentation masks without relying on ground-truth annotations as references at inference time. Given audio-visual-language inputs and each provided segmentation mask, the task requires estimating its IoU with the unobserved ground truth, identifying the corresponding error type, and recommending an actionable quality-control decision. To support this task, we construct MQ-RAVSBench, a benchmark featuring diverse and representative mask error modes that span both geometric and semantic issues. We further propose MQ-Auditor, a multimodal large language model (MLLM)-based auditor that explicitly reasons over multimodal cues and mask information to produce quantitative and qualitative mask quality assessments. Extensive experiments demonstrate that MQ-Auditor outperforms strong open-source and commercial MLLMs and can be integrated with existing Ref-AVS systems to detect segmentation failures and support downstream segmentation improvement. Data and codes will be released at https://github.com/jasongief/MQA-RefAVS.

Jinxing Zhou, Yanghao Zhou, Mingfei Han et al.

Referring Audio-Visual Segmentation (Ref-AVS) aims to segment target objects in audible videos based on given reference expressions. Prior works typically rely on learning latent embeddings via multimodal fusion to prompt a tunable SAM/SAM2 decoder for segmentation, which requires strong pixel-level supervision and lacks interpretability. From a novel perspective of explicit reference understanding, we propose TGS-Agent, which decomposes the task into a Think-Ground-Segment process, mimicking the human reasoning procedure by first identifying the referred object through multimodal analysis, followed by coarse-grained grounding and precise segmentation. To this end, we first propose Ref-Thinker, a multimodal language model capable of reasoning over textual, visual, and auditory cues. We construct an instruction-tuning dataset with explicit object-aware think-answer chains for Ref-Thinker fine-tuning. The object description inferred by Ref-Thinker is used as an explicit prompt for Grounding-DINO and SAM2, which perform grounding and segmentation without relying on pixel-level supervision. Additionally, we introduce R\textsuperscript{2}-AVSBench, a new benchmark with linguistically diverse and reasoning-intensive references for better evaluating model generalization. Our approach achieves state-of-the-art results on both standard Ref-AVSBench and proposed R\textsuperscript{2}-AVSBench. Code will be available at https://github.com/jasongief/TGS-Agent.

Chengzhi Li, Heyan Huang, Ping Jian et al.

Audio-Visual Semantic Segmentation (AVSS) aligns audio and video at the pixel level but requires costly per-frame annotations. We introduce Weakly Supervised Audio-Visual Semantic Segmentation (WSAVSS), which uses only video-level labels to generate per-frame semantic masks of sounding objects. We decompose WSAVSS into looking, listening, and segmentation, and propose Progressive Cross-modal Alignment for Semantics (PCAS) with two modules: *Looking-before-Listening* and *Listening-before-Segmentation*. PCAS builds a classification task to train the audio-visual encoder using video labels, injects visual semantic prompts to enhance frame-level audio understanding, and then applies progressive contrastive alignment to map audio categories to image regions without mask annotations. Experiments show PCAS achieves state-of-the-art performance among weakly supervised methods on AVS and remains competitive with fully supervised baselines on AVSS, validating its effectiveness.

Yanghao Zhou, Jingyu Ma, Yibo Peng et al.

Humanoid control systems have made significant progress in recent years, yet modeling fluent interaction-rich behavior between a robot, its surrounding environment, and task-relevant objects remains a fundamental challenge. This difficulty arises from the need to jointly capture spatial context, temporal dynamics, robot actions, and task intent at scale, which is a poor match to conventional supervision. We propose ExoActor, a novel framework that leverages the generalization capabilities of large-scale video generation models to address this problem. The key insight in ExoActor is to use third-person video generation as a unified interface for modeling interaction dynamics. Given a task instruction and scene context, ExoActor synthesizes plausible execution processes that implicitly encode coordinated interactions between robot, environment, and objects. Such video output is then transformed into executable humanoid behaviors through a pipeline that estimates human motion and executes it via a general motion controller, yielding a task-conditioned behavior sequence. To validate the proposed framework, we implement it as an end-to-end system and demonstrate its generalization to new scenarios without additional real-world data collection. Furthermore, we conclude by discussing limitations of the current implementation and outlining promising directions for future research, illustrating how ExoActor provides a scalable approach to modeling interaction-rich humanoid behaviors, potentially opening a new avenue for generative models to advance general-purpose humanoid intelligence.

Jinxing Zhou, Zhihui Li, Yongqiang Yu et al.

We present \textbf{Met}a-\textbf{T}oken \textbf{Le}arning (Mettle), a simple and memory-efficient method for adapting large-scale pretrained transformer models to downstream audio-visual tasks. Instead of sequentially modifying the output feature distribution of the transformer backbone, Mettle utilizes a lightweight \textit{Layer-Centric Distillation (LCD)} module to distill in parallel the intact audio or visual features embedded by each transformer layer into compact meta-tokens. This distillation process considers both pretrained knowledge preservation and task-specific adaptation. The obtained meta-tokens can be directly applied to classification tasks, such as audio-visual event localization and audio-visual video parsing. To further support fine-grained segmentation tasks, such as audio-visual segmentation, we introduce a \textit{Meta-Token Injection (MTI)} module, which utilizes the audio and visual meta-tokens distilled from the top transformer layer to guide feature adaptation in earlier layers. Extensive experiments on multiple audiovisual benchmarks demonstrate that our method significantly reduces memory usage and training time while maintaining parameter efficiency and competitive accuracy.

Jinxing Zhou, Ziheng Zhou, Yanghao Zhou et al.

The Dense Audio-Visual Event Localization (DAVEL) task aims to temporally localize events in untrimmed videos that occur simultaneously in both the audio and visual modalities. This paper explores DAVEL under a new and more challenging weakly-supervised setting (W-DAVEL task), where only video-level event labels are provided and the temporal boundaries of each event are unknown. We address W-DAVEL by exploiting \textit{cross-modal salient anchors}, which are defined as reliable timestamps that are well predicted under weak supervision and exhibit highly consistent event semantics across audio and visual modalities. Specifically, we propose a \textit{Mutual Event Agreement Evaluation} module, which generates an agreement score by measuring the discrepancy between the predicted audio and visual event classes. Then, the agreement score is utilized in a \textit{Cross-modal Salient Anchor Identification} module, which identifies the audio and visual anchor features through global-video and local temporal window identification mechanisms. The anchor features after multimodal integration are fed into an \textit{Anchor-based Temporal Propagation} module to enhance event semantic encoding in the original temporal audio and visual features, facilitating better temporal localization under weak supervision. We establish benchmarks for W-DAVEL on both the UnAV-100 and ActivityNet1.3 datasets. Extensive experiments demonstrate that our method achieves state-of-the-art performance.

Shuanghong Huang, Jinlei Xu, Youchao Zhou et al.

Large Audio-Language Models (LALMs) have demonstrated strong performance in spoken question answering (QA), with existing evaluations primarily focusing on answer accuracy and robustness to acoustic perturbations. However, such evaluations implicitly assume that spoken inputs remain semantically answerable, an assumption that often fails in real-world interaction when essential information is missing. In this work, we introduce a repair-aware evaluation setting that explicitly distinguishes between answerable and unanswerable audio inputs. We define answerability as a property of the input itself and construct paired evaluation conditions using a semantic-acoustic masking protocol. Based on this setting, we propose the Evaluability Awareness and Repair (EAR) score, a non-compensatory metric that jointly evaluates task competence under answerable conditions and repair behavior under unanswerable conditions. Experiments on two spoken QA benchmarks across diverse LALMs reveal a consistent gap between answer accuracy and conversational reliability: while many models perform well when inputs are answerable, most fail to recognize semantic unanswerability and initiate appropriate conversational repair. These findings expose a limitation of prevailing accuracy-centric evaluation practices and motivate reliability assessments that treat unanswerable inputs as cues for repair and continued interaction.

Dian Jin, Yanghao Zhou, Jinxing Zhou et al.

Referring Audio-Visual Segmentation (Ref-AVS) aims to segment specific objects in videos based on natural language expressions involving audio, vision, and text information. This task poses significant challenges in cross-modal reasoning and fine-grained object localization. In this paper, we propose a simple framework, SimToken, that integrates a multimodal large language model (MLLM) with the Segment Anything Model (SAM). The MLLM is guided to generate a special semantic token representing the referred object. This compact token, enriched with contextual information from all modalities, acts as a prompt to guide SAM to segment objectsacross video frames. To further improve semantic learning, we introduce a novel target-consistent semantic alignment loss that aligns token embeddings from different expressions but referring to the same object. Experiments on the Ref-AVS benchmark demonstrate that our approach achieves superior performance compared to existing methods.

Shixun Huang, Zhifeng Bao, Guoliang Li et al.

Network embedding is an effective method to learn low-dimensional representations of nodes, which can be applied to various real-life applications such as visualization, node classification, and link prediction. Although significant progress has been made on this problem in recent years, several important challenges remain, such as how to properly capture temporal information in evolving networks. In practice, most networks are continually evolving. Some networks only add new edges or nodes such as authorship networks, while others support removal of nodes or edges such as internet data routing. If patterns exist in the changes of the network structure, we can better understand the relationships between nodes and the evolution of the network, which can be further leveraged to learn node representations with more meaningful information. In this paper, we propose the Embedding via Historical Neighborhoods Aggregation (EHNA) algorithm. More specifically, we first propose a temporal random walk that can identify relevant nodes in historical neighborhoods which have impact on edge formations. Then we apply a deep learning model which uses a custom attention mechanism to induce node embeddings that directly capture temporal information in the underlying feature representation. We perform extensive experiments on a range of real-world datasets, and the results demonstrate the effectiveness of our new approach in the network reconstruction task and the link prediction task.