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...
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.
Recent advancements in MLLM-based long-form video understanding have mitigated inference-time computational cost and limited context lengths by selecting query-relevant frames. However, existing approaches predominantly rely on external proxy scorers and rigid heuristic rules, inevitably suffering from misalignment with the target MLLM's intrinsic evidence and failing to accommodate the non-uniform spatiotemporal information density.
The paper introduces Intrinsic Temporal Adaptation (ITA) for Partially Relevant Video Retrieval (PRVR), a task that seeks untrimmed videos containing moments relevant to a text query. ITA employs a Backbone-Internal Temporal Adaptation that lets the final visual transformer layers attend to neighboring frames, creating temporally aware embeddings while keeping CLIP frozen. Additionally, an Affinity-Weighted Gradient Propagation technique softly aggregates top‑k frames based on text‑frame affinities to better handle the weakly supervised nature of PRVR, leading to state‑of‑the‑art performance and more accurate frame‑level evidence retrieval.
Long-video understanding remains challenging for multimodal large language models, because temporally extended videos often contain thousands of frames and are therefore expensive to process exhaustively. Existing methods usually construct compact visual inputs from long videos under a limited visual budget.
CoFiE introduces a two‑stage evidence selection framework for streaming video understanding, separating a coarse, query‑agnostic filtering of visually distinctive frames from a fine, query‑specific refinement during LLM prefill. By filtering out redundant frames before expensive vision encoding, CoFiE reduces end‑to‑end latency while maintaining high accuracy. The method achieves state‑of‑the‑art performance on benchmarks such as StreamingBench and OvO‑Bench, improving accuracy by up to 3.15% and inference speed by up to 2.54× compared to prior approaches.
The paper introduces VT-Contrast, a representation-level temporal counterfactual objective designed to improve temporal understanding in Video Language Models (VideoLMs). By supervising late-layer last-frame video tokens and contrasting order-preserving views with reordered counterfactuals graded by Kendall tau distance, VT-Contrast addresses the mismatch between ordered video input and text-based supervision. The method requires no architectural changes, is compatible with various VideoLM training tasks, and demonstrates improved performance on temporal understanding benchmarks.
TAME introduces a Temporal-Aware Mixture-of-Experts framework for Text-Video Retrieval that enhances CLIP-based models by incorporating frame-level structure and temporal relations. It adds sparse Mixture-of-Experts layers with frame-consistent routing, Frame-Temporal tokens for global cross-frame aggregation, and a Cross-Temporal Interaction and Aggregation module to refine sentence-video similarities. Experiments on multiple TVR benchmarks show consistent performance gains, such as a 4.0 R@1 improvement on MSR‑VTT over CLIP4Clip.
arXiv:2607. 25266v2 Announce Type: replace-cross Abstract: Multimodal large language models (MLLMs) have enabled long-form video understanding at a scale that was not previously possible.
arXiv:2504. 01407v3 Announce Type: replace-cross Abstract: Long video understanding poses a fundamental challenge for large video-language models (LVLMs) due to the overwhelming number of frames and the risk of losing essential context through naive downsampling.
TAME is a CLIP‑based framework for Text‑Video Retrieval that incorporates temporal modeling through three key innovations: sparse Mixture‑of‑Experts layers with frame‑consistent routing, Frame‑Temporal tokens that aggregate cross‑frame information, and a Cross‑Temporal Interaction and Aggregation module for refining frame‑wise similarities. These components enable the model to capture both local visual patterns and long‑range temporal dependencies, leading to consistent performance gains over CLIP‑based baselines on multiple TVR benchmarks, including a 4.0 R@1 improvement on MSR‑VTT. The code is publicly available on GitHub.
arXiv:2605. 21028v2 Announce Type: replace-cross Abstract: Autoregressive long video generation often adopts bounded-memory streaming for efficiency, typically combining local windows for short-term continuity with static early-frame sinks as long-range anchors.