# Chat with AI: The Surprising Turn of Real-time Video Communication from Human to AI

arXiv: https://arxiv.org/abs/2507.10510 · HTML: https://mlsanity.com/paper/2507.10510/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 22 of its 47 references that are in the map.
- First posted on arXiv 14 Jul 2025.

## Builds on (most cited first, 22 of 22)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Neural Discrete Representation Learning](https://mlsanity.com/paper/1711.00937/) (2017)
- [GPT-4o System Card](https://mlsanity.com/paper/2410.21276/) (2024)
- [Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context](https://mlsanity.com/paper/2403.05530/) (2024)
- [Chameleon: Mixed-Modal Early-Fusion Foundation Models](https://mlsanity.com/paper/2405.09818/) (2024)
- [Language Model Beats Diffusion -- Tokenizer is Key to Visual Generation](https://mlsanity.com/paper/2310.05737/) (2023)
- [GLM-4.5V and GLM-4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learning](https://mlsanity.com/paper/2507.01006/) (2025)
- [VideoLLM-online: Online Video Large Language Model for Streaming Video](https://mlsanity.com/paper/2406.11816/) (2024)
- [VITA-1.5: Towards GPT-4o Level Real-Time Vision and Speech Interaction](https://mlsanity.com/paper/2501.01957/) (2025)
- [StreamingBench: Assessing the Gap for MLLMs to Achieve Streaming Video Understanding](https://mlsanity.com/paper/2411.03628/) (2024)
- [Dispider: Enabling Video LLMs with Active Real-Time Interaction via Disentangled Perception, Decision, and Reaction](https://mlsanity.com/paper/2501.03218/) (2025)
- [TimeChat-Online: 80% Visual Tokens are Naturally Redundant in Streaming Videos](https://mlsanity.com/paper/2504.17343/) (2025)
- [MobileCLIP: Fast Image-Text Models through Multi-Modal Reinforced Training](https://mlsanity.com/paper/2311.17049/) (2023)
- [Streaming Video Understanding and Multi-round Interaction with Memory-enhanced Knowledge](https://mlsanity.com/paper/2501.13468/) (2025)
- [StreamBridge: Turning Your Offline Video Large Language Model into a Proactive Streaming Assistant](https://mlsanity.com/paper/2505.05467/) (2025)
- [SVBench: A Benchmark with Temporal Multi-Turn Dialogues for Streaming Video Understanding](https://mlsanity.com/paper/2502.10810/) (2025)
- [OmniMMI: A Comprehensive Multi-modal Interaction Benchmark in Streaming Video Contexts](https://mlsanity.com/paper/2503.22952/) (2025)
- [InternLM-XComposer2.5-OmniLive: A Comprehensive Multimodal System for Long-term Streaming Video and Audio Interactions](https://mlsanity.com/paper/2412.09596/) (2024)
- [AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning](https://mlsanity.com/paper/2412.03248/) (2024)
- [ModServe: Modality- and Stage-Aware Resource Disaggregation for Scalable Multimodal Model Serving](https://mlsanity.com/paper/2502.00937/) (2025)
- [GRACE: Loss-Resilient Real-Time Video through Neural Codecs](https://mlsanity.com/paper/2305.12333/) (2023)
- [Reparo: Loss-Resilient Generative Codec for Video Conferencing](https://mlsanity.com/paper/2305.14135/) (2023)

## Built on by (most cited first, 2 of 2)

- [StreamingAssistant: Efficient Visual Token Pruning for Accelerating Online Video Understanding](https://mlsanity.com/paper/2512.12560/) (2025)
- [Artic: AI-oriented Real-time Communication for MLLM Video Assistant](https://mlsanity.com/paper/2602.12641/) (2026)

## Abstract (arXiv)

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, transmission latency becomes a critical bottleneck preventing AI from being like a real person. To address this, we call for AI-oriented RTC research, exploring the network requirement shift from "humans watching video" to "AI understanding video". We begin by recognizing the main differences between AI Video Chat and traditional RTC. Then, through prototype measurements, we identify that ultra-low bitrate is a key factor for low latency. 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 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. DeViBench is open-sourced at: https://github.com/pku-netvideo/DeViBench.