- Rating
- 1
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper targets KV cache management for streaming VideoQA. The authors observe that existing retrieval methods retrieve only during question prefill or decoding, while visual prefill still suffers from unbounded GPU memory growth or lossy compression. They propose LifeKV, a training-free framework that keeps the full historical visual KV in RAM and recalls query-relevant blocks to the GPU across visual prefill, question prefill, and decoding. Experiments on OVO-Bench and StreamingBench with four backbones show competitive accuracy and lower peak GPU memory.
Critical strengths
-The problem is well motivated and practically important. Extending retrieval to the visual-prefill stage is a sensible fix to a real limitation of prior streaming pipelines.
-The efficiency results, especially the flattened memory growth and latency at long contexts, are encouraging. The training-free design is appealing.
Critical weaknesses
-My main concern is novelty and depth. Each component—KV offloading, top-k retrieval, layer-wise reuse, and prefetching—is familiar from long-context LLM systems; the contribution is mostly a careful combination for VideoQA.
-The evaluation is also narrow: two streaming benchmarks, mostly real-time subsets, and a limited set of baselines. Stronger KV offload/retrieval systems are not compared. Accuracy gains are small and inconsistent across backbones. In some settings LifeKV is close to or below Full Attention, so the claim of competitive accuracy needs more support.
Questions for authors
Strenthen the experimental evaluation. Please inlcude more benchmarks and more recent KV offloading/retrieval baselines, and report accuracy and efficiency under the same controlled setting with more backbone scales.
3 comments
Q1. Expanded Benchmarks, Baselines, and Controlled Evaluation
We have expanded our evaluation with an additional long-video benchmark, more backbone scales, a recent streaming KV baseline, and controlled efficiency comparisons. The additional experiments are summarized below.
(1) Additional Long-Video Benchmark
We evaluate LifeKV, Full Attention, ReKV, and MuKV on RVS-Movie, which contains 22 movie videos averaging approximately one hour. All methods use the same evaluation judge (GLM-5.3-Flash) to ensure consistent scoring.
| Benchmark | Method | Accuracy (%) |
|---|---|---|
| RVS-Movie | Full Attention | 26.51 |
| LifeKV | 25.40 | |
| ReKV | 25.11 | |
| MuKV | 22.53 |
LifeKV achieves accuracy close to Full Attention and outperforms both ReKV and MuKV, supporting its applicability to longer video streams.
(2) Additional Backbone Families and Scales
We extend the evaluation to InternVL2.5-1B and InternVL2.5-4B, supplementing the original LLaVA-OV and Qwen3-VL model families.
| Backbone | OVO-Bench (%) | StreamingBench (%) |
|---|---|---|
| InternVL2.5-1B | 46.91 | 68.14 |
| InternVL2.5-4B | 55.10 | 72.51 |
These results further demonstrate that LifeKV can be applied across different backbone architectures and parameter scales without additional training.
(3) Comparison with a Recent Streaming KV Baseline
We additionally compare LifeKV with StreamKV (AAAI 2026) on OVO-Bench and StreamingBench using LLaVA-OV-0.5B and Qwen3-VL-2B.
| Backbone | Benchmark | Full Attention | StreamKV | LifeKV |
|---|---|---|---|---|
| LLaVA-OV-0.5B | OVO-Bench | 44.35 | 43.14 | 43.58 |
| StreamingBench | 55.11 | 52.51 | 54.67 | |
| Qwen3-VL-2B | OVO-Bench | 53.55 | 53.52 | 54.63 |
| StreamingBench | 71.42 | 71.68 | 70.58 |
LifeKV outperforms StreamKV in three of the four settings, while StreamKV achieves higher accuracy on Qwen3-VL-2B with StreamingBench. Overall, LifeKV maintains accuracy close to Full Attention across the evaluated configurations.
(4) Controlled System Efficiency Comparison
We further compare Full Attention, MuKV, ReKV, StreamKV, and LifeKV on Qwen3-VL-2B using a single NVIDIA A800 GPU. We evaluate video contexts of 128, 512, and 1024 frames and report end-to-end inference latency, peak GPU memory, and host RAM usage.
| Frames | Method | E2E Latency (s) ↓ | Peak VRAM (GiB) ↓ | Host RAM (GiB) ↓ |
|---|---|---|---|---|
| 128 | Full Attention | 2.53 | 7.97 | 2.10 |
| MuKV | 4.21 | 9.50 | 3.70 | |
| ReKV | 3.97 | 9.48 | 2.17 | |
| StreamKV | 5.47 | 5.91 | 2.14 | |
| LifeKV | 4.74 | 4.98 | 5.73 | |
| 512 | Full Attention | 14.80 | 24.61 | 2.60 |
| MuKV | 22.92 | 73.29 | 9.43 | |
| ReKV | 37.05 | 20.66 | 8.04 | |
| StreamKV | 33.25 | 16.02 | 2.71 | |
| LifeKV | 17.18 | 5.07 | 17.53 | |
| 1024 | Full Attention | 47.12 | 53.64 | 3.31 |
| MuKV | 67.20 | 78.50 | 17.05 | |
| ReKV | 156.43 | 48.62 | 15.94 | |
| StreamKV | 103.04 | 39.16 | 3.35 | |
| LifeKV | 36.10 | 5.14 | 32.74 |
LifeKV exhibits the strongest efficiency advantages at longer context lengths. At 1024 frames, it achieves approximately 23.4% lower E2E latency and 90.4% lower peak VRAM than Full Attention, while also being faster than all evaluated retrieval baselines.
Together, these experiments provide broader evidence that LifeKV maintains competitive accuracy across different streaming VideoQA benchmarks and backbones while offering a favorable latency and GPU memory tradeoff for long video streams.
Thanks for your response. I keep my score, due to the limited technical novelty.
W1. Novelty and Technical Contributions
We thank the reviewer for the assessment of novelty and technical depth. We acknowledge that KV offloading, Top-K retrieval, layer-wise reuse, and prefetching have been explored in long-context LLM systems. Our contribution does not lie in these individual components, but in designing a unified full-lifecycle KV retrieval framework tailored to streaming VideoQA.
Unlike conventional long-context inference, streaming VideoQA continuously receives visual inputs without knowing future queries or their required historical evidence. LifeKV addresses this challenge by enabling retrieval across visual prefilling, question prefilling, and decoding, allowing historical visual information to be revisited throughout inference. It further integrates full-fidelity historical KV storage with layer-ahead asynchronous prefetching to reduce GPU memory consumption and retrieval overhead.
We will clarify that the primary contribution is the coordinated design of these mechanisms for scalable streaming inference, rather than the individual techniques themselves. The existing module ablations and system analysis further support their respective contributions.
W2. Evaluation Scope and Accuracy Tradeoff
We acknowledge that the evaluation can be broadened. Our focus is streaming rather than offline VideoQA, where video contexts grow continuously and future queries are unknown. This setting makes persistent historical KV management particularly important. The main paper emphasizes real-time tasks, while the complete OVO-Bench evaluation is already reported in the appendix.
Regarding accuracy, our objective is not to consistently outperform Full Attention, but to preserve its accuracy while substantially reducing the growing GPU memory and inference costs. Performance differences across backbones and tasks remain, and we do not claim uniform improvements. We therefore position competitive accuracy in terms of the tradeoff against efficient streaming KV retrieval methods, rather than absolute superiority over Full Attention.
To strengthen this evidence, we have expanded the evaluation with longer videos, additional backbones, and a recent retrieval baseline, as detailed in Q1.