arXiv:2602. 07739v2 Announce Type: replace-cross Abstract: Embedding geometry plays a fundamental role in retrieval quality, yet dense retrievers for retrieval-augmented generation (RAG) remain largely confined to Euclidean space.
arXiv:2609.36815v1 Announce Type: new Abstract: Partially Relevant Video Retrieval (PRVR) seeks to retrieve untrim-med videos containing a moment that matches a text query, without temporal annotatio...
The paper introduces a retrieval-augmented online textual memory for episodic memory video question answering, focusing on long egocentric videos. It evaluates how different retrieval units, query formulations, context budgets, and reranking strategies affect multiple-choice answering on QAEgo4D-Closed, keeping the generated memory fixed. The best configuration, Parent, matches strong baselines while exposing only a small fraction of the textual memory, and the study shows that better temporal coverage does not always improve accuracy, highlighting the need for evidence-aware representations.
arXiv:2608. 07663v1 Announce Type: cross Abstract: When videos extend from hours to days, directly processing them end-to-end becomes impractical for current Multi-modal Large Language Models (MLLMs).
arXiv:2608.23011v1 Announce Type: cross Abstract: Graph-based retrieval-augmented generation (RAG) provides a scalable paradigm for long-video understanding, but existing systems typically inherit a...
arXiv:2610.08674v1 Announce Type: new Abstract: Current large video-language models (LVLMs) still face challenges when dealing with long videos, mainly because frames are often processed independentl...
Event-Causal RAG (EC‑RAG) is a lightweight retrieval‑augmented framework designed for reasoning over ultra‑long and streaming videos. It segments video streams into semantically complete events using a dual visual‑audio sentinel mechanism, representing each event as a State‑Event‑State (SES) structure that captures pre‑event, event, and post‑event states. During question answering, bidirectional graph retrieval accesses relevant predecessor and successor events from a dual vector‑graph memory, and answers are generated using both this structured memory and the corresponding video evidence. The authors also introduce ECV‑1H, an hour‑scale long‑video QA benchmark with over 150 hours of untrimmed video and 1,251 human‑annotated QA pairs, where EC‑RAG achieves significant accuracy gains across multiple video foundation models while maintaining efficient streaming memory usage on a single RTX 5090 GPU.
arXiv:2608.05592v2 Announce Type: replace Abstract: Multimodal Large Language Models (MLLMs) have made strong progress in video understanding, yet long videos remain difficult: the visual token budge...
arXiv:2607. 24794v1 Announce Type: new Abstract: While Multimodal Large Language Models (MLLMs) demonstrate superior generalization in fundamental video tasks, restricted context windows limit their long video understanding.
arXiv:2608.27871v1 Announce Type: new Abstract: Long-video understanding remains challenging for Multimodal Large Language Models (MLLMs) due to limited context length. Uniform sampling may miss cruc...
Multimodal Large Language Models (MLLMs) have achieved strong progress in video understanding, yet it remains challenging because the token limitation makes MLLMs difficult to capture temporally sparse evidence. Existing methods typically rely on uniform sampling, or frame selection, but these strategies usually optimize either broad temporal coverage or local relevance, making it difficult to preserve both global storyline context and fine-grained evidence.
The paper introduces AURORA, a hyperbolic vision‑language framework that separates angular alignment from radial ordering. Angular alignment captures semantic similarity while a novel Radial‑Angular Order Cone (RAOC) models hierarchical entailment based on parent‑child radial gaps. The approach includes norm regularization to prevent radial degeneration and is evaluated on large image‑text datasets, outperforming both Euclidean and hyperbolic baselines in zero‑shot classification, retrieval, hierarchical reasoning, and compositional understanding.