Pith. sign in

REVIEW 4 cited by

Q-VLM: Post-training Quantization for Large Vision-Language Models

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2410.08119 v3 pith:Q67PPB4L submitted 2024-10-10 cs.CV

classification cs.CV
keywords quantizationdependencysearchcross-layerdiscretizationerrorscostvision-language
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

In this paper, we propose a post-training quantization framework of large vision-language models (LVLMs) for efficient multi-modal inference. Conventional quantization methods sequentially search the layer-wise rounding functions by minimizing activation discretization errors, which fails to acquire optimal quantization strategy without considering cross-layer dependency. On the contrary, we mine the cross-layer dependency that significantly influences discretization errors of the entire vision-language model, and embed this dependency into optimal quantization strategy searching with low search cost. Specifically, we observe the strong correlation between the activation entropy and the cross-layer dependency concerning output discretization errors. Therefore, we employ the entropy as the proxy to partition blocks optimally, which aims to achieve satisfying trade-offs between discretization errors and the search cost. Moreover, we optimize the visual encoder to disentangle the cross-layer dependency for fine-grained decomposition of search space, so that the search cost is further reduced without harming the quantization accuracy. Experimental results demonstrate that our method compresses the memory by 2.78x and increase generate speed by 1.44x about 13B LLaVA model without performance degradation on diverse multi-modal reasoning tasks. Code is available at https://github.com/ChangyuanWang17/QVLM.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 4 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. LatentLLM: Attention-Aware Joint Tensor Compression

    cs.LG 2025-05 conditional novelty 6.0 of 10

    LatentLLM compresses pretrained LLMs and multimodal models with attention-aware joint low-rank tensor decomposition, outperforming SVD-based baselines on OPT perplexity and LLaVA ScienceQA.

  2. LaViDa: A Large Diffusion Language Model for Multimodal Understanding

    cs.CV 2025-05 conditional novelty 6.0 of 10

    A diffusion-based vision-language model matches several autoregressive baselines on multimodal benchmarks while enabling controllable generation and faster decoding at reduced quality.

  3. Speculative Decoding Reimagined for Multimodal Large Language Models

    cs.CV 2025-05 conditional novelty 6.0 of 10

    MSD decouples text and visual tokens in the draft model and trains it first on text, then on a gradually increasing mix of visual data, yielding roughly 2x lossless speedups on LLaVA benchmarks.

  4. $\mu$-MoE: Test-Time Pruning as Micro-Grained Mixture-of-Experts

    cs.LG 2025-05 conditional novelty 4.0 of 10

    Test-time Wanda pruning, reframed as a mixture of micro-experts, adapts the sparse weight mask to each prompt and improves perplexity and VQA accuracy over static pruning baselines.

Pith tools