- Rating
- 1
- Confidence
- 4
- Informativeness
- 2
- Presentation
- 3
- Soundness
- 1
Read review
Paper summary
This paper studies which KV states need to remain after a Video LLM has fully processed a video. It hypothesizes that sparse temporal-carrier positions associated with each model’s temporal serialization are particularly important for subsequent temporal queries, and proposes TC-KV, a query-independent post-prefill compression method that preserves these carriers together with a small visual residual. The method is evaluated on three Video LLMs across temporal grounding, moment retrieval, and highlight detection, with additional interventions analyzing carrier topology and content.
Critical strengths
-
The paper studies a practically relevant post-prefill KV compression setting, where one compact video cache can be reused across multiple queries without re-encoding the video. This is a useful and underexplored efficiency setting.
-
The paper is generally well written and clearly organized. The method and experimental setup are presented in a way that makes the overall argument easy to follow.
Critical weaknesses
-
The central temporal-carrier hypothesis is not strongly supported. Carrier positions are defined a priori from model-specific temporal serialization, and the experiments mainly show that these hand-selected positions are useful retention targets. This does not fully establish the stronger claim that they form a privileged persistent-memory mechanism, especially since other contextualized structural positions are not adequately controlled for.
-
The empirical scope is too limited for the claimed generality. The strongest mechanistic evidence is concentrated on Qwen3-VL, while effects are substantially weaker on the other models. The evaluation also covers a narrow set of temporal tasks and limited baselines, so I am not convinced that the proposed carrier-centric mechanism generalizes broadly across Video LLMs or beyond the specific retention heuristics studied.
Questions for authors
-
What evidence supports treating the designated temporal-carrier positions as a privileged memory mechanism rather than simply useful contextualized structural tokens?
-
How do you reconcile the general carrier-memory claim with the much weaker carrier-content effects observed for LLaVA-OneVision-1.5 and InternVL3.5 compared with Qwen3-VL?