AI Video Chat emerges as a new paradigm for Real-time Communication (RTC), where one peer is not a human, but a Multimodal Large Language Model (MLLM). This makes interaction between humans and AI more intuitive, as if chatting face-to-face with a real person. However, this poses significant challenges to latency, because the MLLM inference takes up most of the response time, leaving very little time for video streaming. Due to network uncertainty and instability, transmission latency becomes a critical bottleneck preventing AI from being like a real person. To address this, we propose Artic, an AI-oriented Real-time Communication framework, exploring the network requirement shift from "humans watching video" to "AI understanding video". To reduce bitrate dramatically while maintaining MLLM accuracy, we propose Context-Aware Video Streaming that recognizes the importance of each video region for chat and allocates bitrate almost exclusively to chat-important regions. To avoid packet retransmission, we propose Loss-Resilient Adaptive Frame Rate that leverages previous frames to substitute for lost/delayed frames while avoiding bitrate waste. To evaluate the impact of video streaming quality on MLLM accuracy, we build the first benchmark, named Degraded Video Understanding Benchmark (DeViBench). Finally, we discuss some open questions and ongoing solutions for AI Video Chat.
Accurate preoperative assessment of lymph node (LN) metastasis in rectal cancer guides treatment decisions, yet conventional MRI evaluation based on morphological criteria shows limited diagnostic performance. While some artificial intelligence models have been developed, they often operate as black boxes, lacking the interpretability needed for clinical trust. Moreover, these models typically evaluate nodes in isolation, overlooking the patient-level context. To address these limitations, we introduce LRMR, an LLM-Driven Relational Multi-node Ranking framework. This approach reframes the diagnostic task from a direct classification problem into a structured reasoning and ranking process. The LRMR framework operates in two stages. First, a multimodal large language model (LLM) analyzes a composite montage image of all LNs from a patient, generating a structured report that details ten distinct radiological features. Second, a text-based LLM performs pairwise comparisons of these reports between different patients, establishing a relative risk ranking based on the severity and number of adverse features. We evaluated our method on a retrospective cohort of 117 rectal cancer patients. LRMR achieved an area under the curve (AUC) of 0.7917 and an F1-score of 0.7200, outperforming a range of deep learning baselines, including ResNet50 (AUC 0.7708). Ablation studies confirmed the value of our two main contributions: removing the relational ranking stage or the structured prompting stage led to a significant performance drop, with AUCs falling to 0.6875 and 0.6458, respectively. Our work demonstrates that decoupling visual perception from cognitive reasoning through a two-stage LLM framework offers a powerful, interpretable, and effective new paradigm for assessing lymph node metastasis in rectal cancer.
Smoke is the first visible indicator of a wildfire.With the advancement of deep learning, image-based smoke detection has become a crucial method for detecting and preventing forest fires. However, the scarcity of smoke image data from forest fires is one of the significant factors hindering the detection of forest fire smoke. Image generation models offer a promising solution for synthesizing realistic smoke images. However, current inpainting models exhibit limitations in generating high-quality smoke representations, particularly manifesting as inconsistencies between synthesized smoke and background contexts. To solve these problems, we proposed a comprehensive framework for generating forest fire smoke images. Firstly, we employed the pre-trained segmentation model and the multimodal model to obtain smoke masks and image captions.Then, to address the insufficient utilization of masks and masked images by inpainting models, we introduced a network architecture guided by mask and masked image features. We also proposed a new loss function, the mask random difference loss, which enhances the consistency of the generated effects around the mask by randomly expanding and eroding the mask edges.Finally, to generate a smoke image dataset using random masks for subsequent detection tasks, we incorporated smoke characteristics and use a multimodal large language model as a filtering tool to select diverse and reasonable smoke images, thereby improving the quality of the synthetic dataset. Experiments showed that our generated smoke images are realistic and diverse, and effectively enhance the performance of forest fire smoke detection models. Code is available at https://github.com/wghr123/MFGDiffusion.
The emergence of Multimodal Large Language Models (MLLMs) has revolutionized image understanding by bridging textual and visual modalities. However, these models often struggle with capturing fine-grained semantic information, such as the precise identification and analysis of object keypoints. Keypoints, as structure-aware, pixel-level, and compact representations of objects, particularly articulated ones, play a crucial role in applications such as fine-grained image analysis, object retrieval, and behavior recognition. In this paper, we propose KptLLM++, a novel multimodal large language model that specifically designed for generic keypoint comprehension through the integration of diverse input modalities guided by user-defined instructions. By unifying keypoint detection across varied contexts, KptLLM++ establishes itself as an advanced interface, fostering more effective human-AI collaboration. The model is built upon a novel identify-then-detect paradigm, which first interprets keypoint semantics and subsequently localizes their precise positions through a structured chain-of-thought reasoning mechanism. To push the boundaries of performance, we have scaled up the training dataset to over 500K samples, encompassing diverse objects, keypoint categories, image styles, and scenarios with complex occlusions. This extensive scaling enables KptLLM++ to unlock its potential, achieving remarkable accuracy and generalization. Comprehensive experiments on multiple keypoint detection benchmarks demonstrate its state-of-the-art performance, underscoring its potential as a unified solution for fine-grained image understanding and its transformative implications for human-AI interaction.
We present BlueLM-2.5-3B, a compact and unified dense Multimodal Large Language Model (MLLM) designed for efficient edge-device deployment, offering strong general-purpose and reasoning capabilities. To the best of our knowledge, this is the first 3B-scale MLLM to support both thinking and non-thinking modes, while also enabling explicit control over thinking token budget. BlueLM-2.5-3B is developed through diversified data curation, key data resampling, hybrid heterogeneous reinforcement learning, and a high-performance training infrastructure. Our model achieves superior multimodal capacity while preserving competitive pure-text performance with only 2.9 billion parameters. We conduct comprehensive evaluations across a broad range of multimodal and text-only benchmarks. In thinking mode, BlueLM-2.5-3B achieves comparable performance to Qwen3-4B on text-only benchmarks, and trails the larger Kimi-VL-A3B-16B by only about 5% on average across multimodal evaluations. In non-thinking mode, it outperforms Qwen2.5-VL-3B on the majority of multimodal benchmarks. Additionally, BlueLM-2.5-3B exhibits exceptional data efficiency. All of the aforementioned performance is achieved with substantially less total training data than Qwen2.5-VL-3B and Qwen3-4B. We hope our work contributes to the advancement of high-performance, on-device MLLMs and provides meaningful insights to the research community.
Virtual try-on has made significant progress in recent years. This paper addresses how to achieve multifunctional virtual try-on guided solely by text instructions, including full outfit change and local editing. Previous methods primarily relied on end-to-end networks to perform single try-on tasks, lacking versatility and flexibility. We propose TalkFashion, an intelligent try-on assistant that leverages the powerful comprehension capabilities of large language models to analyze user instructions and determine which task to execute, thereby activating different processing pipelines accordingly. Additionally, we introduce an instruction-based local repainting model that eliminates the need for users to manually provide masks. With the help of multi-modal models, this approach achieves fully automated local editings, enhancing the flexibility of editing tasks. The experimental results demonstrate better semantic consistency and visual quality compared to the current methods.
Current text-to-image (T2I) generation models struggle to align spatial composition with the input text, especially in complex scenes. Even layout-based approaches yield suboptimal spatial control, as their generation process is decoupled from layout planning, making it difficult to refine the layout during synthesis. We present CoT-Diff, a framework that brings step-by-step CoT-style reasoning into T2I generation by tightly integrating Multimodal Large Language Model (MLLM)-driven 3D layout planning with the diffusion process. CoT-Diff enables layout-aware reasoning inline within a single diffusion round: at each denoising step, the MLLM evaluates intermediate predictions, dynamically updates the 3D scene layout, and continuously guides the generation process. The updated layout is converted into semantic conditions and depth maps, which are fused into the diffusion model via a condition-aware attention mechanism, enabling precise spatial control and semantic injection. Experiments on 3D Scene benchmarks show that CoT-Diff significantly improves spatial alignment and compositional fidelity, and outperforms the state-of-the-art method by 34.7% in complex scene spatial accuracy, thereby validating the effectiveness of this entangled generation paradigm.
Text-to-image retrieval (TIR) aims to find relevant images based on a textual query, but existing approaches are primarily based on whole-image captions and lack interpretability. Meanwhile, referring expression segmentation (RES) enables precise object localization based on natural language descriptions but is computationally expensive when applied across large image collections. To bridge this gap, we introduce Mask-aware TIR (MaTIR), a new task that unifies TIR and RES, requiring both efficient image search and accurate object segmentation. To address this task, we propose a two-stage framework, comprising a first stage for segmentation-aware image retrieval and a second stage for reranking and object grounding with a multimodal large language model (MLLM). We leverage SAM 2 to generate object masks and Alpha-CLIP to extract region-level embeddings offline at first, enabling effective and scalable online retrieval. Secondly, MLLM is used to refine retrieval rankings and generate bounding boxes, which are matched to segmentation masks. We evaluate our approach on COCO and D$^3$ datasets, demonstrating significant improvements in both retrieval accuracy and segmentation quality over previous methods.
Perpetual view generation aims to synthesize a long-term video corresponding to an arbitrary camera trajectory solely from a single input image. Recent methods commonly utilize a pre-trained text-to-image diffusion model to synthesize new content of previously unseen regions along camera movement. However, the underlying 2D diffusion model lacks 3D awareness and results in distorted artifacts. Moreover, they are limited to generating views of static 3D scenes, neglecting to capture object movements within the dynamic 4D world. To alleviate these issues, we present DreamJourney, a two-stage framework that leverages the world simulation capacity of video diffusion models to trigger a new perpetual scene view generation task with both camera movements and object dynamics. Specifically, in stage I, DreamJourney first lifts the input image to 3D point cloud and renders a sequence of partial images from a specific camera trajectory. A video diffusion model is then utilized as generative prior to complete the missing regions and enhance visual coherence across the sequence, producing a cross-view consistent video adheres to the 3D scene and camera trajectory. Meanwhile, we introduce two simple yet effective strategies (early stopping and view padding) to further stabilize the generation process and improve visual quality. Next, in stage II, DreamJourney leverages a multimodal large language model to produce a text prompt describing object movements in current view, and uses video diffusion model to animate current view with object movements. Stage I and II are repeated recurrently, enabling perpetual dynamic scene view generation. Extensive experiments demonstrate the superiority of our DreamJourney over state-of-the-art methods both quantitatively and qualitatively. Our project page: https://dream-journey.vercel.app.
This paper explores the relationship between the condition number of a neural network's weight tensor and the extent of information encoded by the associated processing unit, viewed through the lens of information theory. We argue that a high condition number, though not sufficient for effective knowledge encoding, may indicate that the unit has learned to selectively amplify and compress information. We formalize this intuition, particularly for linear units with Gaussian inputs, linking the condition number and the transformation's log-volume scaling factor to the characteristics of the output entropy and the geometric properties of the learned transformation. Our analysis demonstrates that for a fixed weight norm, a concentrated distribution of singular values (high condition number) corresponds to reduced overall information transfer, indicating a specialized and efficient encoding strategy. Furthermore, we present a practical case study where these principles are applied to guide selective fine-tuning of a multimodal Large Language Model, aiming to mitigate catastrophic forgetting during cross-modal adaptation. Unlike many existing catastrophic forgetting mitigation methods that rely on access to pre-training statistics, which are often unavailable, our selective fine-tuning approach offers a way to bypass this common requirement.
Recent advances in multimodal generative models have unlocked photorealistic, instruction-aligned image generation, yet leading systems like GPT-4o-Image remain proprietary and inaccessible. To democratize these capabilities, we present ShareGPT-4o-Image, the first dataset comprising 45K text-to-image and 46K text-and-image-to-image data, all synthesized using GPT-4o's image generation capabilities for distilling its advanced image generation abilities. Leveraging this dataset, we develop Janus-4o, a multimodal large language model capable of both text-to-image and text-and-image-to-image generation. Janus-4o not only significantly improves text-to-image generation over its predecessor, Janus-Pro, but also newly supports text-and-image-to-image generation. Notably, it achieves impressive performance in text-and-image-to-image generation from scratch, using only 91K synthetic samples and 6 hours of training on an 8 A800-GPU machine. We hope the release of ShareGPT-4o-Image and Janus-4o will foster open research in photorealistic, instruction-aligned image generation.
As one of the earliest ancient languages, Oracle Bone Script (OBS) encapsulates the cultural records and intellectual expressions of ancient civilizations. Despite the discovery of approximately 4,500 OBS characters, only about 1,600 have been deciphered. The remaining undeciphered ones, with their complex structure and abstract imagery, pose significant challenges for interpretation. To address these challenges, this paper proposes a novel two-stage semantic typography framework, named OracleFusion. In the first stage, this approach leverages the Multimodal Large Language Model (MLLM) with enhanced Spatial Awareness Reasoning (SAR) to analyze the glyph structure of the OBS character and perform visual localization of key components. In the second stage, we introduce Oracle Structural Vector Fusion (OSVF), incorporating glyph structure constraints and glyph maintenance constraints to ensure the accurate generation of semantically enriched vector fonts. This approach preserves the objective integrity of the glyph structure, offering visually enhanced representations that assist experts in deciphering OBS. Extensive qualitative and quantitative experiments demonstrate that OracleFusion outperforms state-of-the-art baseline models in terms of semantics, visual appeal, and glyph maintenance, significantly enhancing both readability and aesthetic quality. Furthermore, OracleFusion provides expert-like insights on unseen oracle characters, making it a valuable tool for advancing the decipherment of OBS.
Accurate and interpretable multi-disease diagnosis remains a critical challenge in medical research, particularly when leveraging heterogeneous multimodal medical data. Current approaches often rely on single-modal data, limiting their ability to comprehensively understand complex diseases. To address this, we propose MedTVT-R1, a novel Multimodal Large Language Model (MLLM) framework designed to integrate clinical multimodal data for reasoning and diagnosing multiple diseases. We construct MedTVT-QA, a curated instruction dataset that provides question-answer pairs for physiological-level interpretations and disease-level diagnoses with a Chain of Evidence approach. MedTVT-R1 incorporates a modality perception layer to capture inter-modal dependencies and adaptively weight modality contributions. Additionally, we employ Group Relative Policy Optimization (GRPO)-based Reinforcement Fine-Tuning with a Jaccard Reward function to enhance diagnostic reasoning. Experimental results demonstrate MedTVT-R1's superiority in multimodal feature utilization and multi-disease diagnosis, offering significant potential for clinical applications such as diagnostic report generation and comorbidity reasoning. The dataset and code are available at https://github.com/keke-nice/MedTVT-R1.
While end-to-end video-to-audio generation has greatly improved, producing high-fidelity audio that authentically captures the nuances of visual content remains challenging. Like professionals in the creative industries, such generation requires sophisticated reasoning about items such as visual dynamics, acoustic environments, and temporal relationships. We present \textbf{ThinkSound}, a novel framework that leverages Chain-of-Thought (CoT) reasoning to enable stepwise, interactive audio generation and editing for videos. Our approach decomposes the process into three complementary stages: foundational foley generation that creates semantically coherent soundscapes, interactive object-centric refinement through precise user interactions, and targeted editing guided by natural language instructions. At each stage, a multimodal large language model generates contextually aligned CoT reasoning that guides a unified audio foundation model. Furthermore, we introduce \textbf{AudioCoT}, a comprehensive dataset with structured reasoning annotations that establishes connections between visual content, textual descriptions, and sound synthesis. Experiments demonstrate that ThinkSound achieves state-of-the-art performance in video-to-audio generation across both audio metrics and CoT metrics and excels in out-of-distribution Movie Gen Audio benchmark. The demo page is available at https://ThinkSound-Demo.github.io.
Despite remarkable advancements in text-to-image person re-identification (TIReID) facilitated by the breakthrough of cross-modal embedding models, existing methods often struggle to distinguish challenging candidate images due to intrinsic limitations, such as network architecture and data quality. To address these issues, we propose an Interactive Cross-modal Learning framework (ICL), which leverages human-centered interaction to enhance the discriminability of text queries through external multimodal knowledge. To achieve this, we propose a plug-and-play Test-time Humane-centered Interaction (THI) module, which performs visual question answering focused on human characteristics, facilitating multi-round interactions with a multimodal large language model (MLLM) to align query intent with latent target images. Specifically, THI refines user queries based on the MLLM responses to reduce the gap to the best-matching images, thereby boosting ranking accuracy. Additionally, to address the limitation of low-quality training texts, we introduce a novel Reorganization Data Augmentation (RDA) strategy based on information enrichment and diversity enhancement to enhance query discriminability by enriching, decomposing, and reorganizing person descriptions. Extensive experiments on four TIReID benchmarks, i.e., CUHK-PEDES, ICFG-PEDES, RSTPReid, and UFine6926, demonstrate that our method achieves remarkable performance with substantial improvement.
Multimodal Large Language Models (MLLMs) have rapidly evolved with the growth of Large Language Models (LLMs) and are now applied in various fields. In finance, the integration of diverse modalities such as text, charts, and tables is crucial for accurate and efficient decision-making. Therefore, an effective evaluation system that incorporates these data types is essential for advancing financial application. In this paper, we introduce CFBenchmark-MM, a Chinese multimodal financial benchmark with over 9,000 image-question pairs featuring tables, histogram charts, line charts, pie charts, and structural diagrams. Additionally, we develop a staged evaluation system to assess MLLMs in handling multimodal information by providing different visual content step by step. Despite MLLMs having inherent financial knowledge, experimental results still show limited efficiency and robustness in handling multimodal financial context. Further analysis on incorrect responses reveals the misinterpretation of visual content and the misunderstanding of financial concepts are the primary issues. Our research validates the significant, yet underexploited, potential of MLLMs in financial analysis, highlighting the need for further development and domain-specific optimization to encourage the enhanced use in financial domain.
In the field of multimodal chain-of-thought (CoT) reasoning, existing approaches predominantly rely on reasoning on pure language space, which inherently suffers from language bias and is largely confined to math or science domains. This narrow focus limits their ability to handle complex visual reasoning tasks that demand comprehensive understanding of image details. To address these limitations, this paper introduces VGR, a novel reasoning multimodal large language model (MLLM) with enhanced fine-grained visual perception capabilities. Unlike traditional MLLMs that answer the question or reasoning solely on the language space, our VGR first detects relevant regions that may help to solve problems, and then provides precise answers based on replayed image regions. To achieve this, we conduct a large-scale SFT dataset called VGR -SFT that contains reasoning data with mixed vision grounding and language deduction. The inference pipeline of VGR allows the model to choose bounding boxes for visual reference and a replay stage is introduced to integrates the corresponding regions into the reasoning process, enhancing multimodel comprehension. Experiments on the LLaVA-NeXT-7B baseline show that VGR achieves superior performance on multi-modal benchmarks requiring comprehensive image detail understanding. Compared to the baseline, VGR uses only 30\% of the image token count while delivering scores of +4.1 on MMStar, +7.1 on AI2D, and a +12.9 improvement on ChartQA.
Multimodal Large Language Model (MLLM) often suffer from hallucinations. They over-rely on partial cues and generate incorrect responses. Recently, methods like Visual Contrastive Decoding (VCD) and Instruction Contrastive Decoding (ICD) have been proposed to mitigate hallucinations by contrasting predictions from perturbed or negatively prefixed inputs against original outputs. In this work, we uncover that methods like VCD and ICD fundamentally influence internal attention dynamics of the model. This observation suggests that their effectiveness may not stem merely from surface-level modifications to logits but from deeper shifts in attention distribution. Inspired by this insight, we propose an attention-steerable contrastive decoding framework that directly intervenes in attention mechanisms of the model to offer a more principled approach to mitigating hallucinations. Our experiments across multiple MLLM architectures and diverse decoding methods demonstrate that our approach significantly reduces hallucinations and improves the performance on benchmarks such as POPE, CHAIR, and MMHal-Bench, while simultaneously enhancing performance on standard VQA benchmarks.
Data visualization generation using Large Language Models (LLMs) has shown promising results but often produces suboptimal visualizations that require human intervention for improvement. In this work, we introduce VIS-Shepherd, a specialized Multimodal Large Language Model (MLLM)-based critic to evaluate and provide feedback for LLM-generated data visualizations. At the core of our approach is a framework to construct a high-quality visualization critique dataset, where we collect human-created visualization instances, synthesize corresponding LLM-generated instances, and construct high-quality critiques. We conduct both model-based automatic evaluation and human preference studies to evaluate the effectiveness of our approach. Our experiments show that even small (7B parameters) open-source MLLM models achieve substantial performance gains by leveraging our high-quality visualization critique dataset, reaching levels comparable to much larger open-source or even proprietary models. Our work demonstrates significant potential for MLLM-based automated visualization critique and indicates promising directions for enhancing LLM-based data visualization generation. Our project page: https://github.com/bopan3/VIS-Shepherd.
Phrases are essential to understand the core concepts in conversations. However, due to their rare occurrence in training data, correct translation of phrases is challenging in speech translation tasks. In this paper, we propose a phrase dictionary biasing method to leverage pairs of phrases mapping from the source language to the target language. We apply the phrase dictionary biasing method to two types of widely adopted models, a transducer-based streaming speech translation model and a multimodal large language model. Experimental results show that the phrase dictionary biasing method outperforms phrase list biasing by 21% relatively for the streaming speech translation model. In addition, phrase dictionary biasing enables multimodal large language models to use external phrase information, achieving 85% relative improvement in phrase recall.