InternVideo3: Agentify Foundation Models with Multimodal Contextual Reasoning

• URL: http://arxiv.org/abs/2606.12195v1
• Date: Wed, 10 Jun 2026 15:17:08 GMT
• Title: InternVideo3: Agentify Foundation Models with Multimodal Contextual Reasoning
• Abstract summary: We present InternVideo3, a framework enhancing multimodal understanding in video tasks.<n>MCR treats understanding as a closed-loop process over a shared, evolving context.<n>We show InternVideo3 achieves strong performance on benchmarks like Video-MME, MLVU, and Ego.
• Score: 61.87294123934038
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recent progress in foundation models has shifted toward agentic behavior involving multi-step reasoning and tool use. However, open-source efforts largely focus on text-dominant settings, leaving long-horizon multimodal tasks underexplored. This gap is evident in video tasks requiring sustained temporal understanding and iterative interaction. We present InternVideo3, a framework enhancing these capabilities via Multimodal Contextual Reasoning (MCR). MCR treats understanding as a closed-loop process over a shared, evolving context containing observations, instructions, reasoning, tool actions, and memory. This frames long-video understanding as evidence accumulation and verification. To ensure efficiency, we introduce Multimodal Multi-head Latent Attention (M^2LA), a token-preserving reparameterization compressing KV-cache states while retaining the full token stream. Our staged training includes continued pretraining, short-to-long supervised fine-tuning, rule-based reinforcement learning, and on-policy distillation. Experiments show InternVideo3 achieves strong performance on benchmarks like Video-MME, MLVU, and EgoSchema. We further instantiate the model as a video agent with retrieval tools, demonstrating robust evidence-grounded behavior. Our results suggest that efficient context handling and closed-loop reasoning are vital for adapting open multimodal models toward long-horizon visually grounded agency.

Related papers

• VideoScout: Learning Agentic Active Exploration with Adaptive Reasoning Pacing for Long Video Understanding [61.00809248898572]
  Long video understanding is a problem in which an agent reads the video turn by turn along the temporal axis, deciding at each turn how fast to watch, what evidence to retain, when to revisit uncertain segments, and when to stop and answer.<n>We propose textbfVideoScout, a multi-turn reasoning agent that instantiates the SEA paradigm through adaptive reasoning pacing.<n>Experiments on long video understanding and reasoning benchmarks demonstrate that our 7B model achieves strong performance compared with existing trained 7B agentic models.
  arXiv  Detail & Related papers  (2026-09-14T14:04:03Z)
• STORM: Internalized Modeling for Spatial-Temporal Reasoning in Video-Language Models [81.32710031596591]
  Video reasoning tasks require tracking motion, temporal order, and evolving visual states across frames.<n>Existing methods built on vision-language models (LVLMs) often address this challenge by externalizing reasoning through chain-of-thought (CoT)<n>We propose STORMS, a two-stage framework that teaches LVLMs to reason through bounded continuous latent trajectories instead of explicit textual CoT.
  arXiv  Detail & Related papers  (2026-05-25T16:33:00Z)
• Video-MME-v2: Towards the Next Stage in Benchmarks for Comprehensive Video Understanding [98.3098451637867]
  Video-MME-v2 is a comprehensive benchmark designed to rigorously evaluate the robustness and faithfulness of video understanding.<n>To guarantee data quality, Video-MME-v2 is constructed through a rigorously controlled human annotation pipeline.
  arXiv  Detail & Related papers  (2026-04-06T17:59:56Z)
• EventMemAgent: Hierarchical Event-Centric Memory for Online Video Understanding with Adaptive Tool Use [9.106278853322008]
  EventMemAgent is an active online video agent framework based on a hierarchical memory module.<n>Our framework employs a dual-layer strategy for online videos: short-term memory detects event boundaries and utilizes event-granular reservoir sampling to process streaming video frames.<n>Experiments show that EventMemAgent internalizes competitive results on online video benchmarks.
  arXiv  Detail & Related papers  (2026-02-17T03:17:24Z)
• VideoZoomer: Reinforcement-Learned Temporal Focusing for Long Video Reasoning [49.35834435935727]
  VideoZoomer is a novel agentic framework that enables MLLMs to control their visual focus during reasoning.<n>Our 7B model delivers diverse and complex reasoning patterns, yielding strong performance across a broad set of long video understanding and reasoning benchmarks.<n>These emergent capabilities allow it to consistently surpass existing open-source models and even rival proprietary systems on challenging tasks.
  arXiv  Detail & Related papers  (2025-12-26T11:43:21Z)
• Video-LMM Post-Training: A Deep Dive into Video Reasoning with Large Multimodal Models [78.32948112203228]
  Video understanding represents the most challenging frontier in computer vision.<n>Recent emergence of Video-Large Multitemporal Models has demonstrated remarkable capabilities in video understanding tasks.<n>Survey aims to provide researchers and practitioners with a unified framework for advancing Video-LMM capabilities.
  arXiv  Detail & Related papers  (2025-10-06T17:10:44Z)
• Bisecle: Binding and Separation in Continual Learning for Video Language Understanding [11.710573955384511]
  We propose Bisecle for video-language continual learning, inspired by the rapid Binding and pattern separation mechanisms in the hippocampus.<n>Bisecle mitigates forgetting and enhances cross-task generalization on several VideoQA benchmarks.
  arXiv  Detail & Related papers  (2025-07-01T06:28:57Z)
• DaMO: A Data-Efficient Multimodal Orchestrator for Temporal Reasoning with Video LLMs [5.074812070492738]
  We introduce DaMO, a data-efficient Video LLM specifically designed for accurate temporal reasoning and multimodal understanding.<n>We train DaMO via a structured four-stage progressive training paradigm, incrementally equipping the model with multimodal alignment, semantic grounding, and temporal reasoning capabilities.<n>Our work establishes a promising direction for data-efficient video-language modeling.
  arXiv  Detail & Related papers  (2025-06-13T08:13:05Z)

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.