通过对比反事实评估优化视觉令牌传输，提升重建质量而不增加发送数据量。
Baseline-Relative Counterfactual Refinement for Bit-Aware Visual Token Communication
- 基于局部模型的反事实修正机制，动态评估令牌选择优劣
- 在低中速率下显著提升重建质量，且不增加实际传输包率
- 适用于多种数据集、信道和分辨率，适合资源受限场景
生成式视觉令牌通信通过仅发送选定的离散令牌并由接收端重构缺失内容来降低传输负载。然而，现有基于局部不确定性、重要性或多样性的令牌选择标准无法直接判断在相同包预算下改变当前选择是否能提升最终重建质量。为此，我们提出面向通信的门控反事实精炼（GCR-C），作为对局部-最小描述长度（Local-MDL）的滚动式修正层。GCR-C构建紧凑多样的候选令牌集，通过匹配全预算的Local-MDL延续过程评估每个候选方案，并仅在获得正向基线相对重建增益时替换原始选择。在CIFAR-10、STL-10、编码5G-LDPC链路及高分辨率Kodak传输数据集上的实验表明，GCR-C在活跃的低中速率工作点上持续提升重建质量，且不增加实际包率，同时对数据集、信道条件、分辨率、令牌网格与分词器的变化保持有效。结果还揭示了因额外编码端反事实评估带来的清晰质量-计算权衡。
原文摘要 · Abstract (English)
Generative visual-token communication reduces transmission load by sending only selected discrete tokens and reconstructing missing content at the receiver. However, existing token-selection criteria based on local uncertainty, importance, or diversity do not directly determine whether changing the current selection improves the final reconstruction under the same packet budget. To address this problem, we propose Gated Counterfactual Refinement for Communication (GCR-C), a rollout-style correction layer over Local-MDL. GCR-C constructs a compact diversified candidate set, evaluates each candidate through matched full-budget Local-MDL continuation, and replaces the baseline action only when a positive baseline-relative reconstruction gain is obtained. Experiments on CIFAR-10, STL-10, a coded 5G-LDPC link, and a limited high-resolution Kodak transfer show that GCR-C consistently improves reconstruction quality at active low- and medium-rate operating points without increasing the realized packet rate, while remaining effective across changes in dataset, channel condition, resolution, token grid, and tokenizer. The results also reveal a clear quality--computation tradeoff due to the additional encoder-side counterfactual evaluation.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。