mlsanityData to 14 Oct 2026

LLMC+: Benchmarking Vision-Language Model Compression with a Plug-and-play Toolkit

arXiv 2508.09981 ↗First posted 13 Aug 2025cs.CVcited by 7

On the map: Multimodal LLMs / vision-language models → Visual token pruning and efficiency

Abstract (arXiv)

Large Vision-Language Models (VLMs) exhibit impressive multi-modal capabilities but suffer from prohibitive computational and memory demands, due to their long visual token sequences and massive parameter sizes. To address these issues, recent works have proposed training-free compression methods. However, existing efforts often suffer from three major limitations: (1) Current approaches do not decompose techniques into comparable modules, hindering fair evaluation across spatial and temporal redundancy. (2) Evaluation confined to simple single-turn tasks, failing to reflect performance in realistic scenarios. (3) Isolated use of individual compression techniques, without exploring their joint potential. To overcome these gaps, we introduce LLMC+, a comprehensive VLM compression benchmark with a versatile, plug-and-play toolkit. LLMC+ supports over 20 algorithms across five representative VLM families and enables systematic study of token-level and model-level compression. Our benchmark reveals that: (1) Spatial and temporal redundancies demand distinct technical strategies. (2) Token reduction methods degrade significantly in multi-turn dialogue and detail-sensitive tasks. (3) Combining token and model compression achieves extreme compression with minimal performance loss. We believe LLMC+ will facilitate fair evaluation and inspire future research in efficient VLM. Our code is available at https://github.com/ModelTC/LightCompress.

Key facts. Cited by 7 papers in the map (data through 14 Oct 2026). Placed in Visual token pruning and efficiency, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 42 of its 66 references that are in the map. First posted on arXiv 13 Aug 2025.

Builds on

42 of 66 references.

Closest ancestors · 6

Same direction, newest first

  1. 2025Stop Looking for Important Tokens in Multimodal Language Models: Duplication Matters More17 Feb
  2. 2024FrameFusion: Combining Similarity and Importance for Video Token Reduction on Large Vision Language Models30 Dec
  3. 2024Beyond Text-Visual Attention: Exploiting Visual Cues for Effective Token Pruning in VLMs2 Dec
  4. 2024Filter, Correlate, Compress: Training-Free Token Reduction for MLLM Acceleration26 Nov
  5. 2024PyramidDrop: Accelerating Your Large Vision-Language Models via Pyramid Visual Redundancy Reduction22 Oct
  6. 2024SparseVLM: Visual Token Sparsification for Efficient Vision-Language Model Inference6 Oct

Other references · 36

  1. 2020Language Models are Few-Shot Learnerscited by 32,419
  2. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 14,132
  3. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,944
  4. 2025Qwen2.5-VL Technical Reportcited by 5,587
  5. 2023Improved Baselines with Visual Instruction Tuningcited by 4,418
  6. 2024Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolutioncited by 4,234
  7. 2023Sigmoid Loss for Language Image Pre-Trainingcited by 3,109
  8. 2024LLaVA-OneVision: Easy Visual Task Transfercited by 2,719
  9. 2023InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Taskscited by 2,565
  10. 2015VQA: Visual Question Answeringcited by 2,506
  11. 2023MMBench: Is Your Multi-modal Model an All-around Player?cited by 2,141
  12. 2022GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformerscited by 1,919
24 more
  1. 2019Towards VQA Models That Can Readcited by 1,718
  2. 2020Knowledge Distillation: A Surveycited by 1,603
  3. 2024Video-MME: The First-Ever Comprehensive Evaluation Benchmark of Multi-modal LLMs in Video Analysiscited by 1,413
  4. 2023Video-LLaVA: Learning United Visual Representation by Alignment Before Projectioncited by 1,402
  5. 2023AWQ: Activation-aware Weight Quantization for LLM Compression and Accelerationcited by 1,298
  6. 2023SGLang: Efficient Execution of Structured Language Model Programscited by 1,141
  7. 2023SparseGPT: Massive Language Models Can Be Accurately Pruned in One-Shotcited by 1,121
  8. 2023MVBench: A Comprehensive Multi-modal Video Understanding Benchmarkcited by 1,109
  9. 2022Token Merging: Your ViT But Fastercited by 1,021
  10. 2024LongVideoBench: A Benchmark for Long-context Interleaved Video-Language Understandingcited by 665
  11. 2023OCRBench: On the Hidden Mystery of OCR in Large Multimodal Modelscited by 476
  12. 2024GaLore: Memory-Efficient LLM Training by Gradient Low-Rank Projectioncited by 463
  13. 2024SVD-LLM: Truncation-aware Singular Value Decomposition for Large Language Model Compressioncited by 243
  14. 2023TFMQ-DM: Temporal Feature Maintenance Quantization for Diffusion Modelscited by 69
  15. 2024DB-LLM: Accurate Dual-Binarization for Efficient LLMscited by 37
  16. 2024PTQ4SAM: Post-Training Quantization for Segment Anythingcited by 32
  17. 2025QVGen: Pushing the Limit of Quantized Video Generative Modelscited by 22
  18. 2025Past-Future Scheduler for LLM Serving under SLA Guaranteescited by 20
  19. 2020Channel Pruning Guided by Classification Loss and Feature Importancecited by 17
  20. 2024HarmoniCa: Harmonizing Training and Inference for Better Feature Caching in Diffusion Transformer Accelerationcited by 17
  21. 2024QVD: Post-training Quantization for Video Diffusion Modelscited by 14
  22. 2024Temporal Feature Matters: A Framework for Diffusion Model Quantizationcited by 5
  23. 2024PTSBench: A Comprehensive Post-Training Sparsity Benchmark Towards Algorithms and Modelscited by 5
  24. 2025Model Compression using Progressive Channel Pruningcited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 1; 2026: 6.

7 papers.

  1. 2026Light Forcing: Accelerating Autoregressive Video Diffusion via Sparse Attentioncited by 12
  2. 2025MoDES: Accelerating Mixture-of-Experts Multimodal Large Language Models via Dynamic Expert Skippingcited by 11
  3. 2026Focus-dLLM: Accelerating Long-Context Diffusion LLM Inference via Confidence-Guided Context Focusingcited by 7
  4. 2026Why and When Visual Token Pruning Fails? A Study on Relevant Visual Information Shift in MLLMs Decodingcited by 7
  5. 2026M3-ACE: Rectifying Visual Perception in Multimodal Math Reasoning via Multi-Agentic Context Engineeringcited by 1
  6. 2026GoQuant: Geometric Orthogonal Residual Projection for Multiplier-Free Power-of-Two Transformer Quantizationcited by 1
  7. 2026Attribution-Guided and Coverage-Maximized Pruning for Structural MoE Compressioncited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Hierarchical Compression of Vision-Language Model Benchmarks
  2. 2025LVLM-Compress-Bench: Benchmarking the Broader Impact of Large Vision-Language Model Compression
  3. 2025EffiVLM-BENCH: A Comprehensive Benchmark for Evaluating Training-Free Acceleration in Large Vision-Language Models
  4. 2025Hybrid Token Compression for Vision-Language Models
  5. 2025A Glimpse to Compress: Dynamic Visual Token Pruning for Large Vision-Language Models
  6. 2024VL-Cache: Sparsity and Modality-Aware KV Cache Compression for Vision-Language Model Inference Acceleration
  7. 2025Variation-aware Vision Token Dropping for Faster Large Vision-Language Models
  8. 2026Vision Token Reduction via Attention-Driven Self-Compression for Efficient Multimodal Large Language Models