Temporally-Grounded Language Generation: A Benchmark for Real-Time Vision-Language Models

• URL: http://arxiv.org/abs/2505.11326v1
• Date: Fri, 16 May 2025 14:48:30 GMT
• Title: Temporally-Grounded Language Generation: A Benchmark for Real-Time Vision-Language Models
• Abstract summary: Vision-language models (VLMs) have shown remarkable progress in offline tasks such as image captioning and video question answering.<n>Real-time interactive environments impose new demands on VLMs, requiring them to generate utterances that are not only semantically accurate but also precisely timed.<n>We present $textbfVision-Language Model with Time-Synchronized Interleaving (VLM-TSI)$, a model that interleaves visual and linguistic tokens in a time-synchronized manner.
• Score: 14.69446469340837
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Vision-language models (VLMs) have shown remarkable progress in offline tasks such as image captioning and video question answering. However, real-time interactive environments impose new demands on VLMs, requiring them to generate utterances that are not only semantically accurate but also precisely timed. We identify two core capabilities necessary for such settings -- $\textit{perceptual updating}$ and $\textit{contingency awareness}$ -- and propose a new benchmark task, $\textbf{Temporally-Grounded Language Generation (TGLG)}$, to evaluate them. TGLG requires models to generate utterances in response to streaming video such that both content and timing align with dynamic visual input. To support this benchmark, we curate evaluation datasets from sports broadcasting and egocentric human interaction domains, and introduce a new metric, $\textbf{TRACE}$, to evaluate TGLG by jointly measuring semantic similarity and temporal alignment. Finally, we present $\textbf{Vision-Language Model with Time-Synchronized Interleaving (VLM-TSI)}$, a model that interleaves visual and linguistic tokens in a time-synchronized manner, enabling real-time language generation without relying on turn-based assumptions. Experimental results show that VLM-TSI significantly outperforms a strong baseline, yet overall performance remains modest -- highlighting the difficulty of TGLG and motivating further research in real-time VLMs. Code and data available $\href{https://github.com/yukw777/tglg}{here}$.

Related papers

• Mitigating Modality and Language-Style Gaps for Zero-Shot Video Moment Retrieval [41.787320093855506]
  We propose Self-Similarity-based Moment Proposal and Scoring.<n>By deriving self-similarity only from the video content, we circumvent the noisy and mismatched patterns of query-frame or query-caption similarities.<n>We also introduce a query-aware MLLM-based reasoning stage to further sharpen alignment between text and video.
  arXiv  Detail & Related papers  (2026-07-21T12:16:29Z)
• Real-Time Generation of Game Video Commentary with Multimodal LLMs: Pause-Aware Decoding Approaches [69.57389826203699]
  We investigate whether in-context prompting alone can support real-time commentary generation that is both semantically relevant and well-timed.<n>We propose two prompting-based decoding strategies: 1) a fixed-interval approach, and 2) a novel dynamic interval-based decoding approach.<n> Experiments on Japanese and English datasets of racing and fighting games show that the dynamic interval-based decoding can generate commentary more closely aligned with human utterance timing and content using prompting alone.
  arXiv  Detail & Related papers  (2026-03-03T06:39:04Z)
• TV-RAG: A Temporal-aware and Semantic Entropy-Weighted Framework for Long Video Retrieval and Understanding [14.570869250170139]
  TV-RAG is a training-free architecture that couples temporal alignment with entropy-guided semantics to improve long-video reasoning.<n>By weaving these temporal and semantic signals together, TV-RAG realises a dual-level reasoning routine that can be grafted onto any LVLM without re-training or fine-tuning.
  arXiv  Detail & Related papers  (2025-12-29T14:10:22Z)
• TimeLens: Rethinking Video Temporal Grounding with Multimodal LLMs [81.78017865436816]
  We present TimeLens, a systematic investigation into building MLLMs with strong video temporal grounding ability.<n>We first expose critical quality issues in existing VTG benchmarks and introduce TimeLens-Bench, comprising meticulously re-annotated versions of three popular benchmarks.<n>We also address noisy training data through an automated re-annotation pipeline, yielding TimeLens-100K, a large-scale, high-quality training dataset.
  arXiv  Detail & Related papers  (2025-12-16T18:59:58Z)
• Harnessing Synthetic Preference Data for Enhancing Temporal Understanding of Video-LLMs [54.502280390499756]
  We propose TimeWarp to create a targeted synthetic temporal dataset to fine-tune the model's responses to encourage it to focus on the given input video.<n>We demonstrate that when our method is applied to existing models, it significantly improves performance on temporal understanding benchmarks.
  arXiv  Detail & Related papers  (2025-10-04T21:48:40Z)
• Adapting LLMs to Time Series Forecasting via Temporal Heterogeneity Modeling and Semantic Alignment [32.41581846555808]
  Large Language Models (LLMs) have recently demonstrated impressive capabilities in natural language processing.<n>We propose TALON, a unified framework that enhances LLM-based forecasting by modeling temporal and enforcing semantic alignment.<n>Experiments on seven real-world benchmarks demonstrate that TALON achieves superior performance across all datasets.
  arXiv  Detail & Related papers  (2025-08-10T06:06:19Z)
• VideoMolmo: Spatio-Temporal Grounding Meets Pointing [66.19964563104385]
  VideoMolmo is a model tailored for fine-grained pointing of video sequences.<n>A novel temporal mask fusion employs SAM2 for bidirectional point propagation.<n>To evaluate the generalization of VideoMolmo, we introduce VPoMolS-temporal, a challenging out-of-distribution benchmark spanning five real-world scenarios.
  arXiv  Detail & Related papers  (2025-06-05T17:59:29Z)
• DisTime: Distribution-based Time Representation for Video Large Language Models [23.176698643825123]
  DisTime is a lightweight framework designed to enhance temporal comprehension in Video-LLMs.<n>DisTime employs a learnable token to create a continuous temporal embedding space.<n>DisTime achieves state-of-the-art performance across benchmarks in three time-sensitive tasks.
  arXiv  Detail & Related papers  (2025-05-30T08:10:18Z)
• InfLVG: Reinforce Inference-Time Consistent Long Video Generation with GRPO [73.33751812982342]
  InfLVG is an inference-time framework that enables coherent long video generation without requiring additional long-form video data.<n>We show that InfLVG can extend video length by up to 9$times$, achieving strong consistency and semantic fidelity across scenes.
  arXiv  Detail & Related papers  (2025-05-23T07:33:25Z)
• Everything Can Be Described in Words: A Simple Unified Multi-Modal Framework with Semantic and Temporal Alignment [0.0]
  Long Video Question Answering (LVQA) is challenging due to the need for temporal reasoning and large-scale multimodal data processing.<n>We introduce UMaT, a retrieval-augmented generation framework that efficiently processes extremely long videos.<n>We show that UMaT outperforms existing methods in multimodal integration, long-form video understanding, and sparse information retrieval.
  arXiv  Detail & Related papers  (2025-03-12T05:28:24Z)
• Temporal Contrastive Learning for Video Temporal Reasoning in Large Vision-Language Models [44.99833362998488]
  Temporal Semantic Alignment via Dynamic Prompting (TSADP) is a novel framework that enhances temporal reasoning capabilities.<n>We evaluate TSADP on the VidSitu dataset, augmented with enriched temporal annotations.<n>Our analysis highlights the robustness, efficiency, and practical utility of TSADP, making it a step forward in the field of video-language understanding.
  arXiv  Detail & Related papers  (2024-12-16T02:37:58Z)
• Text-Video Retrieval with Global-Local Semantic Consistent Learning [122.15339128463715]
  We propose a simple yet effective method, Global-Local Semantic Consistent Learning (GLSCL) GLSCL capitalizes on latent shared semantics across modalities for text-video retrieval. Our method achieves comparable performance with SOTA as well as being nearly 220 times faster in terms of computational cost.
  arXiv  Detail & Related papers  (2024-05-21T11:59:36Z)
• Temporally Consistent Referring Video Object Segmentation with Hybrid Memory [98.80249255577304]
  We propose an end-to-end R-VOS paradigm that explicitly models temporal consistency alongside the referring segmentation. Features of frames with automatically generated high-quality reference masks are propagated to segment remaining frames. Extensive experiments demonstrate that our approach enhances temporal consistency by a significant margin.
  arXiv  Detail & Related papers  (2024-03-28T13:32:49Z)
• Zero-Shot Video Moment Retrieval from Frozen Vision-Language Models [58.17315970207874]
  We propose a zero-shot method for adapting generalisable visual-textual priors from arbitrary VLM to facilitate moment-text alignment. Experiments conducted on three VMR benchmark datasets demonstrate the notable performance advantages of our zero-shot algorithm.
  arXiv  Detail & Related papers  (2023-09-01T13:06:50Z)
• Fine-grained Semantic Alignment Network for Weakly Supervised Temporal Language Grounding [148.46348699343991]
  Temporal language grounding aims to localize a video segment in an untrimmed video based on a natural language description. Most of the existing weakly supervised methods generate a candidate segment set and learn cross-modal alignment through a MIL-based framework. We propose a novel candidate-free framework: Fine-grained Semantic Alignment Network (FSAN), for weakly supervised TLG.
  arXiv  Detail & Related papers  (2022-10-21T13:10:27Z)
• Self-supervised Learning for Semi-supervised Temporal Language Grounding [84.11582376377471]
  Temporal Language Grounding (TLG) aims to localize temporal boundaries of the segments that contain the specified semantics in an untrimmed video. Previous works either tackle this task in a fully-supervised setting that requires a large amount of manual annotations or in a weakly supervised setting that cannot achieve satisfactory performance. To achieve good performance with limited annotations, we tackle this task in a semi-supervised way and propose a unified Semi-supervised Temporal Language Grounding (STLG) framework.
  arXiv  Detail & Related papers  (2021-09-23T16:29:16Z)

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.