Evaluation of state-of-the-art open-source and closed-source MLLMs reveals that current MLLMs function as lossy summarizers rather than faithful memorizers, highlighting the need for architectures with genuine long-term spatiotemporal memory.
Abstract
Recent advances in Multimodal Large Language Models (MLLMs) have led to substantial progress in video understanding, accompanied by a growing number of long video benchmarks. However, existing benchmarks rely predominantly on web-sourced videos that lack inter-clip spatiotemporal continuity, making it difficult to assess whether models can maintain consistent memory across days or weeks of real-world experience. We introduce EgoMonth, the first month-level egocentric video understanding benchmark. EgoMonth comprises over 300 hours of first-person daily-life recordings from 20 participants spanning 20 to 120 days, paired with 1,443 human-crafted multiple-choice question-answer pairs. We design a cognitively grounded 14-task evaluation framework organized into three hierarchical cognitive levels: Schema Consolidation, Episodic Indexing, and Cascading Reasoning. Evaluation of state-of-the-art open-source and closed-source MLLMs reveals that even the best-performing model, Gemini 2.5 Pro, achieves only 71.8% macro-average accuracy, remaining 22.4 percentage points below the corrected human baseline of 94.2%. Several models perform near or below the 25% chance level on tasks such as Route Reasoning, Cross-view Spatial Reasoning, and Direction Judgement, while even the strongest closed-source model remains substantially below human performance. These results indicate that current MLLMs function as lossy summarizers rather than faithful memorizers, highlighting the need for architectures with genuine long-term spatiotemporal memory.
EgoCross is a cross-domain egocentric video question answering benchmark designed to evaluate whether multimodal large language models can generalize beyond common daily-life scenarios. The first EgoCross Challenge was hosted at the Third EgoVis Workshop at CVPR 2026 and evaluated models on first-person videos from four target domains: surgery, industrial assembly, extreme sports, and animal perspectives. Each test example consists of an egocentric video clip, a question, and four candidate answers, from which the model must select the correct option. This technical report introduces the challenge task, benchmark resources, and two official Codabench tracks. The Source-Limited Track restricts participants to the official baseline model and a small support set, whereas the Open-Source Track permits broader choices of models and training data under rules that prohibit the manual construction of target-domain training data. In total, the challenge received more than 1,500 submissions from over 130 participants, with 19 teams participating in the Open-Source Track and 38 teams in the Source-Limited Track. We further present the official leaderboard results and summarize the winning solutions from both tracks. We hope that this report will serve as a useful technical reference for advancing cross-domain egocentric video understanding. All resources, including the challenge data, baseline implementation, and code released by the winning teams, are made publicly available.
Yuqian Fu, Tianwen Qian, Yanjun Li et al.· 0 citations
Agentic video understanding equips models with long-term memory to autonomously process and respond to continuous, long-horizon multimodal streams. However, advanced video agents often rely on ``detective-style''iterative reasoning for action control (e.g., $\mathtt{search}$) and evidence aggregation, incurring prohibitive costs and latency. We argue that such heavy reasoning primarily compensates for the lack of global context and semantic misalignment in retrieval. This paper introduces Light-Omni, a multimodal agent framework for reflexive and lightweight video understanding. It achieves this through dual contextual states that instantly build the required context in a single forward pass. First, we maintain a global state, a finite-sized multimodal script continuously consolidated from episodic memory, serving as the global context for Light-Omni. Through hierarchical merging, it preserves recent details while summarizing past events. Second, conditioned on this global context, we generate a parametric latent state that directly drives autonomous actions and produces retrieval embeddings, with minimal latency. Benefiting from this coupled design, Light-Omni achieves semantically aligned retrieval and reflexive responses while avoiding iterative reasoning. Extensive experiments validate the effectiveness of Light-Omni across multiple video benchmarks. Notably, it outperforms M3-Agent with an average 2.4% accuracy gain, a 12.1$\times$ speedup, and a 2.6$\times$ improvement in GPU memory efficiency. Furthermore, it serves as a memory system to enhance both the performance and efficiency of existing MLLMs. Project page: https://clare-nie.github.io/Light-Omni.
Chang Nie, Jiaju Wei, Jun-Lan Feng et al.· 0 citations
To facilitate smart wearable devices or human-like robotics with real-time first-person perspective perception ability, recent researchers proposed the Egocentric Online Action Segmentation (EOAS) task. It requires models to recognize what is happening in egocentric streaming videos and discriminate the starting and ending times of an activity in a real-time manner. However, compared with offline-recorded exocentric videos, egocentric streaming videos cannot provide equivalent sufficient temporal-spatial cues due to the limited perspective and unknown coming frames. Hence, it raises a high demand for the long-term episodic memory ability of models. To this end, most previous approaches work on compressing long-term memory into feature representations. In this paper, we propose a novel EOAS paradigm, termed Parametric Context Memory Learning (PCML), which integrates episodic memory into learnable parameters and keeps dynamic updates according to real-time frames. Concretely, we design the Parametric Context Perception layer and construct a novel Episodic Semantic Memorization Network (ESMN) based on it, which integrates episodic memory into learnable parameters and keeps dynamic updates with real-time frames. We evaluate our proposed method on three public egocentric streaming video benchmarks including EgoPER, EgoProceL, and GTEA. Extensive experiments demonstrate the ESMN model significantly outperforms recent state-of-the-art methods. Our code is available at https://github.com/XunCHN/PCML
Xun Jiang, Xing Xu, Chong Liu et al.· IEEE Transactions on Image P...· 1 citation
Building assistants that can continually watch the world, remember what they see, and reason over their accumulated experience is a long-standing goal, and recently multimodal agents equipped with long-term memory over video streams have attracted increasing interest. Unfortunately, existing systems either keep their memory inside the model context or in a flat feature store, and organize it around frames rather than around the persistent entities a stream is really about, which confines them to bounded videos and weakens their ability to track who and what reappears over time. In this paper, we propose ReflectWorld-MM, an entity-oriented multimodal memory system for open-ended video streams. It consists of three parts. The first is a perception front-end that turns an audiovisual stream into entity-resolved observations under a bounded short-term memory. The second is a hierarchical long-term memory, grounded in human memory theory, that couples a multi-scale episodic memory, an evolving entity-centric semantic memory, and a procedural memory. The third is a complete realization, built for real-world operation, that ingests arbitrary streams and plugs into off-the-shelf assistants. Across six long-video and lifelong-memory benchmarks, ReflectWorld-MM achieves the best accuracy on all six, outperforming strong memory agents and a frontier model.
Xiaokang Ma, Yifan Sun, Zhihong Jin et al.· 0 citations
Large vision-language models (LVLMs) have shown significant progress in video understanding, but they struggle to scale to long videos due to limited context windows. Existing methods reduce input dimensionality via frame sampling and feature compression, yet discard details and incur high computational cost for post-training. In contrast, retrieval-augmented generation (RAG) that indexes long videos for query retrieval and memory-based methods that maintain evolving long-term stores, offer a lighter and deployment-friendly solution. Nevertheless, they rely on shallow retrieval that selects only top-ranked segments and fails to integrate information across multiple relevant video episodes. Inspired by Multiple-Trace Theory in cognitive psychology, we revisit long video understanding from a probe-echo perspective, in which human episodic memories are activated and integrated in parallel. Building on this insight, we propose ProEchoMem, a cognitive-inspired framework that simulates the probe-echo mechanism: (1) Incremental Episodic Memory Construction builds structured knowledge graphs from video streams; (2) Probe-Driven Memory Activation generates probe signals from user queries to activate all stored traces simultaneously; (3) Memory Echo Synthesis integrates activated traces into a coherent and structured memory echo. Experiments on LongerVideos, LVBench, and cross-domain settings demonstrate the effectiveness of ProEchoMem, with multi-trace probing achieving up to 14.2% higher relevance and ablation studies validating the contribution of each module. The code is available at https://github.com/Applied-Machine-Learning-Lab/SIGIR26_ProEchoMem
Derong Xu, Yanxin Chen, Wanyu Wang et al.· Annual International ACM SIG...· 0 citations
Video generative models provide rich physical priors for robot learning, yet existing world-action models (WAMs) face a fundamental trade-off: synchronous video generation at control rate is latency-prohibitive, while abandoning test-time visual imagination sacrifices task success. We show that visual imagination achieves both real-time inference and superior success rates when generated asynchronously off the critical path and consumed directly in latent space. We introduce GlanceWAM, which decouples imagination from control within a single video DiT: an asynchronous proposer glances ahead on a slow clock to imagine a single lookahead frame seconds into the future in the background, while an action head decodes action chunks at control rate (48 ms) purely in latent space without blocking. Enabled by a non-interfering attention mask that isolates video representations and staleness-robust horizon training that accommodates asynchronous lookahead aging, GlanceWAM breaks the speed-success dilemma. Trained purely on demonstrations, it attains 72.2% on the 24-task RoboCasa kitchen benchmark (surpassing synchronous Cosmos Policy at 67.1% and imagination-free co-training at 64.4%) and 99.0% on LIBERO, executing at 48 ms per chunk on an NVIDIA A100 GPU (24x faster than synchronous baselines). Code is available at https://github.com/linhanwang/GlanceWAM.
Linhan Wang, Zijian An, Mingyuan Zhang et al.· 0 citations