Baseline-Relative Counterfactual Refinement for Bit-Aware Visual Token Communication
Not on the map. Why?
Abstract (arXiv)
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.
Key facts. Cited by 1 paper in the map (data through 13 Oct 2026). Builds on 34 of its 45 references that are in the map. First posted on arXiv 17 Aug 2026.
Builds on
34 of 45 references.
References · 34
- 2017Neural Discrete Representation Learningcited by 4,680
- 2020Taming Transformers for High-Resolution Image Synthesiscited by 3,237
- 2022MaskGIT: Masked Generative Image Transformercited by 1,090
- 2021DynamicViT: Efficient Vision Transformers with Dynamic Token Sparsificationcited by 852
- 2024Autoregressive Model Beats Diffusion: Llama for Scalable Image Generationcited by 764
- 2018Joint Autoregressive and Hierarchical Priors for Learned Image Compressioncited by 658
- 2020Deep Learning Enabled Semantic Communication Systemscited by 289
- 2025DivPrune: Diversity-based Visual Token Pruning for Large Multimodal Modelscited by 177
- 2022Sionna: An Open-Source Library for Next-Generation Physical Layer Researchcited by 145
- 2024Fit and Prune: Fast and Training-free Visual Token Pruning for Multi-modal Large Language Modelscited by 125
- 2024ATP-LLaVA: Adaptive Token Pruning for Large Vision Language Modelscited by 87
- 2024VoCo-LLaMA: Towards Vision Compression with Large Language Modelscited by 78
22 more
- 2025PACT: Pruning and Clustering-Based Token Reduction for Faster Visual Language Modelscited by 49
- 2024Feather the Throttle: Revisiting Visual Token Pruning for Vision-Language Model Accelerationcited by 39
- 2024Language-Guided Image Tokenization for Generationcited by 36
- 2024Dynamic-VLM: Simple Dynamic Visual Token Compression for VideoLLMcited by 31
- 2024Inference Optimal VLMs Need Fewer Visual Tokens and More Parameterscited by 25
- 2024PVC: Progressive Visual Token Compression for Unified Image and Video Processing in Large Vision-Language Modelscited by 25
- 2024Token Cropr: Faster ViTs for Quite a Few Taskscited by 18
- 2025Keyframe-oriented Vision Token Pruning: Enhancing Efficiency of Large Vision Language Models on Long-Form Video Processingcited by 18
- 2025CROP: Contextual Region-Oriented Visual Token Pruningcited by 14
- 2025Adaptive Semantic Token Communication for Transformer-based Edge Inferencecited by 13
- 2022Generative Model Based Highly Efficient Semantic Communication Approach for Image Transmissioncited by 11
- 2025Skip-Vision: Efficient and Scalable Acceleration of Vision-Language Models via Adaptive Token Skippingcited by 11
- 2025Representation Shift: Unifying Token Compression with FlashAttentioncited by 5
- 2025SQ-GAN: Semantic Image Communications Using Masked Vector Quantizationcited by 4
- 2025Faster Parameter-Efficient Tuning with Token Redundancy Reductioncited by 4
- 2026Energy-Driven Adaptive Visual Token Pruning for Efficient Vision-Language Modelscited by 4
- 2026RTPrune: Reading-Twice Inspired Token Pruning for Efficient DeepSeek-OCR Inferencecited by 4
- 2019Biased Aggregation, Rollout, and Enhanced Policy Improvement for Reinforcement Learningcited by 2
- 2025FCoT-VL:Advancing Text-oriented Large Vision-Language Models with Efficient Visual Token Compressioncited by 2
- 2026World Model-Enabled Causal Digital Twins for Semantic Communications in Physical AI Systemscited by 2
- 2026Visual Token Compression Enhances Robustness of MLLMscited by 2
- 2026Vision Token Reduction via Attention-Driven Self-Compression for Efficient Multimodal Large Language Modelscited by 1
Built on by
Citing papers per year: 2026: 1.
1 paper.