video-SALMONN 2: Caption-Enhanced Audio-Visual Large Language Models

• URL: http://arxiv.org/abs/2506.15220v3
• Date: Fri, 26 Sep 2025 07:30:12 GMT
• Title: video-SALMONN 2: Caption-Enhanced Audio-Visual Large Language Models
• Authors: Changli Tang, Yixuan Li, Yudong Yang, Jimin Zhuang, Guangzhi Sun, Wei Li, Zejun Ma, Chao Zhang,
• Abstract summary: We present video-SALMONN 2, a family of audio-visual large language models that set new state-of-the-art (SOTA) results in video description and question answering (QA)<n>Our core contribution is multi-round direct preference optimisation (MrDPO), paired with a caption-quality objective that jointly rewards completeness and factual accuracy.
• Score: 47.74219861820857
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We present video-SALMONN 2, a family of audio-visual large language models that set new state-of-the-art (SOTA) results in video description and question answering (QA). Our core contribution is multi-round direct preference optimisation (MrDPO), paired with a caption-quality objective that jointly rewards completeness and factual accuracy. Unlike standard DPO with a fixed reference policy, MrDPO periodically refreshes the reference by bootstrapping from a newly re-initialised lightweight adapter trained on the latest preferences, avoiding reference staleness and enabling continual improvement. This strategy produces captions that are consistently more detailed and accurate than those from proprietary systems such as GPT-4o and Gemini-1.5 Pro. We further distil these gains by using our model to generate a high-quality video-caption corpus for supervised fine-tuning of new models, transferring benefits beyond captioning to strong performance on complex video-QA tasks. Across widely used audio-visual and visual-only understanding benchmarks (including Video-MME, WorldSense, AVUT, Video-Holmes, DailyOmni, MLVU, and LVBench), our 3B and 7B models achieve SOTA results at comparable scales, while the 72B model surpasses all other open-source systems. Our source code, models, and data are released at \href{https://github.com/bytedance/video-SALMONN-2}{https://github.com/bytedance/video-SALMONN-2}.

Related papers

• JavisDiT++: Unified Modeling and Optimization for Joint Audio-Video Generation [112.614973927778]
  Joint audio-video generation (JAVG) produces synchronized and semantically aligned sound and vision from textual descriptions.<n>This paper presents JavisDiT++, a framework for unified modeling and optimization of JAVG.<n>Our model achieves state-of-the-art performance merely with around 1M public training entries.
  arXiv  Detail & Related papers  (2026-02-22T12:44:28Z)
• MOVA: Towards Scalable and Synchronized Video-Audio Generation [91.56945636522345]
  We introduce MOVA (MOSS Video and Audio), an open-source model capable of generating high-quality, synchronized audio-visual content.<n>By releasing the model weights and code, we aim to advance research and foster a vibrant community of creators.
  arXiv  Detail & Related papers  (2026-02-09T15:31:54Z)
• ALIVE: Animate Your World with Lifelike Audio-Video Generation [50.693986608051716]
  ALIVE is a generation model that adapts a pretrained Text-to-Video (T2V) model to Sora-style audio-video generation and animation.<n>To support the audio-visual synchronization and reference animation, we augment the popular MMDiT architecture with a joint audio-video branch.<n>ALIVE demonstrates outstanding performance, consistently outperforming open-source models and matching or surpassing state-of-the-art commercial solutions.
  arXiv  Detail & Related papers  (2026-02-09T14:06:03Z)
• LTX-2: Efficient Joint Audio-Visual Foundation Model [3.1804093402153506]
  LTX-2 is an open-source model capable of generating temporally synchronized audiovisual content.<n>We employ a multilingual text encoder for broader prompt understanding.<n>LTX-2 produces rich, coherent audio tracks that follow the characters, environment, style, and emotion of each scene.
  arXiv  Detail & Related papers  (2026-01-06T18:24:41Z)
• UGC-VideoCaptioner: An Omni UGC Video Detail Caption Model and New Benchmarks [13.205921806688147]
  Real-world user-generated videos, especially on platforms like TikTok, often feature rich and intertwined audio visual content.<n>Existing video captioning benchmarks and models remain predominantly visual centric, overlooking the crucial role of audio in conveying scene dynamics, speaker intent, and narrative context.<n>To address these challenges, we introduce-VideoCap, a new benchmark and model framework specifically designed for detailed omnimodal captioning of short form user-generated videos.
  arXiv  Detail & Related papers  (2025-07-15T14:08:29Z)
• AVC-DPO: Aligned Video Captioning via Direct Preference Optimization [50.08618093204503]
  Video multimodal large language models (video MLLMs) have achieved substantial progress in video captioning tasks.<n>We propose Aligned Video Captioning via Direct Preference Optimization (AVC-DPO), a post-training framework designed to enhance captioning capabilities in video MLLMs through preference alignment.<n>We have achieved exceptional performance in the LOVE@PRCV'25 Workshop Track 1A: Video Detailed Captioning Challenge, achieving first place on the Video Detailed Captioning benchmark.
  arXiv  Detail & Related papers  (2025-07-02T08:51:45Z)
• DenseDPO: Fine-Grained Temporal Preference Optimization for Video Diffusion Models [60.716734545171114]
  We introduce DenseDPO, a method that addresses shortcomings by making three contributions.<n>First, we create each video pair for DPO by denoising corrupted copies of a ground truth video.<n>Second, we leverage the resulting temporal alignment to label preferences on short segments rather than entire clips, yielding a denser and more precise learning signal.
  arXiv  Detail & Related papers  (2025-06-04T03:06:08Z)
• SynPO: Synergizing Descriptiveness and Preference Optimization for Video Detailed Captioning [69.34975070207763]
  We leverage preference learning to enhance the performance of vision-language models in fine-grained video captioning.<n>We propose a novel optimization method offering significant advantages over DPO and its variants.<n>Results demonstrate that SynPO consistently outperforms DPO variants while achieving 20% improvement in training efficiency.
  arXiv  Detail & Related papers  (2025-06-01T04:51:49Z)
• VistaDPO: Video Hierarchical Spatial-Temporal Direct Preference Optimization for Large Video Models [80.92928946973026]
  We introduce VistaDPO, a framework for Video Hierarchical Spatial-Temporal Direct Preference Optimization.<n> VistaDPO enhances text-video preference alignment across three hierarchical levels.<n>Experiments on benchmarks such as Video Hallucination, Video QA, and Captioning performance tasks demonstrate that VistaDPO significantly improves the performance of existing LVMs.
  arXiv  Detail & Related papers  (2025-04-17T17:39:41Z)
• video-SALMONN-o1: Reasoning-enhanced Audio-visual Large Language Model [33.70837005629285]
  We propose video-SALMONN-o1, the first open-source reasoning-enhanced audio-visual LLM designed for general video understanding tasks.<n>We develop a reasoning-intensive dataset featuring challenging audio-visual questions with step-by-step solutions.<n>We also introduce RivaBench, the first reasoning-intensive video understanding benchmark, featuring over 4,000 high-quality, expert-curated question-answer pairs.
  arXiv  Detail & Related papers  (2025-02-17T13:07:40Z)
• Temporal Preference Optimization for Long-Form Video Understanding [28.623353303256653]
  Temporal Preference Optimization (TPO) is a novel post-training framework designed to enhance the temporal grounding capabilities of video-LMMs.<n>TPO significantly enhances temporal understanding while reducing reliance on manually annotated data.<n>LLaVA-Video-TPO establishes itself as the leading 7B model on the Video-MME benchmark.
  arXiv  Detail & Related papers  (2025-01-23T18:58:03Z)
• Enhancing Multimodal LLM for Detailed and Accurate Video Captioning using Multi-Round Preference Optimization [19.327911862822262]
  We present video-SALMONN 2, an advanced audio-visual large language model (LLM) with low-rank adaptation (LoRA) We propose new metrics to evaluate the completeness and accuracy of video descriptions, which are optimized using directed preference optimization (DPO) Experiments show that mrDPO significantly enhances video-SALMONN 2's captioning accuracy, reducing global and local error rates by 40% and 20%, respectively.
  arXiv  Detail & Related papers  (2024-10-09T08:44:47Z)
• VideoLLaMA 2: Advancing Spatial-Temporal Modeling and Audio Understanding in Video-LLMs [55.82090875098132]
  VideoLLaMA 2 is a set of Video Large Language Models (Video-LLMs) designed to enhance spatial-temporal modeling and audio understanding in video and audio-oriented tasks. VideoLLaMA 2 consistently achieves competitive results among open-source models and even gets close to some proprietary models on several benchmarks.
  arXiv  Detail & Related papers  (2024-06-11T17:22:23Z)
• Direct Preference Optimization of Video Large Multimodal Models from Language Model Reward [118.65089648651308]
  This paper introduces a novel framework that utilizes detailed video captions as a proxy of video content. We show that applying this tailored reward through DPO significantly improves the performance of video LMMs on video Question Answering (QA) tasks.
  arXiv  Detail & Related papers  (2024-04-01T17:28:16Z)
• Video-Teller: Enhancing Cross-Modal Generation with Fusion and Decoupling [79.49128866877922]
  Video-Teller is a video-language foundation model that leverages multi-modal fusion and fine-grained modality alignment. Video-Teller boosts the training efficiency by utilizing frozen pretrained vision and language modules. It capitalizes on the robust linguistic capabilities of large language models, enabling the generation of both concise and elaborate video descriptions.
  arXiv  Detail & Related papers  (2023-10-08T03:35:27Z)
• HowToCaption: Prompting LLMs to Transform Video Annotations at Scale [72.69268311756082]
  We propose to leverage the capabilities of large language models (LLMs) to obtain high-quality video descriptions aligned with videos at scale. We introduce a prompting method that is able to take into account a longer text of subtitles, allowing us to capture the contextual information beyond one single sentence. We apply our method to the subtitles of the HowTo100M dataset, creating a new large-scale dataset, HowToCaption.
  arXiv  Detail & Related papers  (2023-10-07T19:32:55Z)
• AV-data2vec: Self-supervised Learning of Audio-Visual Speech Representations with Contextualized Target Representations [88.30635799280923]
  We introduce AV-data2vec which builds audio-visual representations based on predicting contextualized representations. Results on LRS3 show that AV-data2vec consistently outperforms existing methods with the same amount of data and model size.
  arXiv  Detail & Related papers  (2023-02-10T02:55:52Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.