EventCoT: Event-centric Video Chain-of-thought for Reasoning Temporal Localization

• URL: http://arxiv.org/abs/2607.04872v1
• Date: Mon, 06 Jul 2026 09:47:23 GMT
• Title: EventCoT: Event-centric Video Chain-of-thought for Reasoning Temporal Localization
• Abstract summary: Reasoning temporal localization (RTL) requires a model to generate an answer that itself contains the time interval supporting it.<n>We propose the first event-centric video chain-of-thought framework, dubbed EventCoT.<n>EventCoT first performs event-centric tokenization of the input video to convert it into compact event tokens.<n>It then reasons within the identified events to generate the answer, grounding the time interval via embedding matching.
• Score: 39.30013113761109
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Reasoning temporal localization (RTL) requires a model to generate an answer that itself contains the time interval supporting it, so high-level reasoning and precise temporal grounding must be produced jointly in a single response. To tackle this challenging task, we propose the first event-centric video chain-of-thought framework, dubbed EventCoT. EventCoT first performs event-centric tokenization of the input video to convert it into compact event tokens, enabling efficient identification of question-relevant events. It then reasons within the identified events to generate the answer, grounding the time interval via embedding matching that aligns placeholder tokens with visual embeddings. EventCoT achieves state-of-the-art results on ActivityNet-RTL for reasoning temporal localization while using substantially fewer visual tokens than previous work. To verify its general performance, we further evaluate EventCoT on the grounded video question answering benchmark ReXTime, where it attains strong zero-shot results.

Related papers

• Event ActivityNet: A Large-Scale Simulated-Event Benchmark for Untrimmed Action Understanding [26.582838811672413]
  Event ActivityNet is a large-scale simulated-event benchmark from human-annotated, untrimmed ActivityNet videos.<n>It supports annotated-segment action recognition, auxiliary event-language alignment, and causal online temporal action localization.<n>Under a progressive nested-scale training protocol, recognition Top-1 accuracy increases from 52.25 to 66.42, while online temporal localization average mAP improves from 21.7 to 29.0.
  arXiv  Detail & Related papers  (2026-08-03T09:20:46Z)
• E.M.Ground: A Temporal Grounding Vid-LLM with Holistic Event Perception and Matching [87.38371267983263]
  Temporal Video Grounding aims to precisely localize time segments corresponding to query events.<n>E.M.Ground is a novel Vid-LLM for TVG that focuses on holistic and coherent event perception.<n>E.M.Ground consistently outperforms state-of-the-art Vid-LLMs by significant margins.
  arXiv  Detail & Related papers  (2026-02-05T02:16:00Z)
• EventSTU: Event-Guided Efficient Spatio-Temporal Understanding for Video Large Language Models [56.16721798968254]
  We propose an event-guided, training-free framework for efficient understanding, named EventSTU.<n>In the temporal domain, we design a coarse-to-fine sampling algorithm that the change-triggered property of event cameras to eliminate redundant large frames.<n>In the spatial domain, we achieves an adaptive token pruning algorithm that leverages the saliency of events as a zero-cost prior to guide spatial reduction.
  arXiv  Detail & Related papers  (2025-11-24T09:30:02Z)
• PR-DETR: Injecting Position and Relation Prior for Dense Video Captioning [23.4119982709261]
  We propose a novel dense video captioning framework, named PR-DETR, which injects the explicit position and relation prior into the detection transformer.<n>On the one hand, we first generate a set of position-anchored queries to provide the scene-specific position and semantic information about potential events as position prior.<n>On the other hand, we further design an event relation encoder to explicitly calculate the relationship between event boundaries as relation prior to guide the event interaction to improve the semantic coherence of the captions.
  arXiv  Detail & Related papers  (2025-06-19T07:07:42Z)
• Uneven Event Modeling for Partially Relevant Video Retrieval [17.05658098065075]
  We propose an Uneven Event Modeling (UEM) framework for partially relevant video retrieval (PRVR)<n>We first introduce the Progressive-Grouped Video (PGVS) module, to iteratively formulate events in light of both temporal dependencies and semantic similarity between consecutive frames.<n>We also propose the Context-Aware Event Refinement (CAER) module to refine the event representation conditioned the text's cross-attention.
  arXiv  Detail & Related papers  (2025-06-01T08:21:45Z)
• EventCrab: Harnessing Frame and Point Synergy for Event-based Action Recognition and Beyond [61.10181853363728]
  Event-based Action Recognition (EAR) possesses the advantages of high-temporal and privacy preservation compared with traditional action recognition.<n>We present EventCrab, a framework that adeptly integrates the "lighter" frame-specific networks for dense event frames with the "heavier" point-specific networks for sparse event points.<n>Experiments on four datasets demonstrate the significant performance of our proposed EventCrab.
  arXiv  Detail & Related papers  (2024-11-27T13:28:57Z)
• Unifying Event Detection and Captioning as Sequence Generation via Pre-Training [53.613265415703815]
  We propose a unified pre-training and fine-tuning framework to enhance the inter-task association between event detection and captioning. Our model outperforms the state-of-the-art methods, and can be further boosted when pre-trained on extra large-scale video-text data.
  arXiv  Detail & Related papers  (2022-07-18T14:18:13Z)
• Event Transformer [43.193463048148374]
  Event camera's low power consumption and ability to capture microsecond brightness make it attractive for various computer vision tasks. Existing event representation methods typically convert events into frames, voxel grids, or spikes for deep neural networks (DNNs) This work introduces a novel token-based event representation, where each event is considered a fundamental processing unit termed an event-token.
  arXiv  Detail & Related papers  (2022-04-11T15:05:06Z)
• Team RUC_AIM3 Technical Report at Activitynet 2020 Task 2: Exploring Sequential Events Detection for Dense Video Captioning [63.91369308085091]
  We propose a novel and simple model for event sequence generation and explore temporal relationships of the event sequence in the video. The proposed model omits inefficient two-stage proposal generation and directly generates event boundaries conditioned on bi-directional temporal dependency in one pass. The overall system achieves state-of-the-art performance on the dense-captioning events in video task with 9.894 METEOR score on the challenge testing set.
  arXiv  Detail & Related papers  (2020-06-14T13:21:37Z)

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.