Pith. sign in

REVIEW 5 cited by

Vision as LoRA

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

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 2503.20680 v1 pith:RQO7OHEK submitted 2025-03-26 cs.CV cs.CL

Han Wang , Yongjie Ye , Bingru Li , Yuxiang Nie , Jinghui Lu , Jingqun Tang , Yanjie Wang , Can Huang This is my paper
classification cs.CVcs.CL
keywords voraloravisionvisualcapabilitiescontextdataintroduce
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

We introduce Vision as LoRA (VoRA), a novel paradigm for transforming an LLM into an MLLM. Unlike prevalent MLLM architectures that rely on external vision modules for vision encoding, VoRA internalizes visual capabilities by integrating vision-specific LoRA layers directly into the LLM. This design allows the added parameters to be seamlessly merged into the LLM during inference, eliminating structural complexity and minimizing computational overhead. Moreover, inheriting the LLM's ability of handling flexible context, VoRA can process inputs at arbitrary resolutions. To further strengthen VoRA's visual capabilities, we introduce a block-wise distillation method that transfers visual priors from a pre-trained ViT into the LoRA layers, effectively accelerating training by injecting visual knowledge. Additionally, we apply bi-directional attention masks to better capture the context information of an image. We successfully demonstrate that with additional pre-training data, VoRA can perform comparably with conventional encode-based MLLMs. All training data, codes, and model weights will be released at https://github.com/Hon-Wong/VoRA.

Discussion (0). Sign in to comment.

Forward citations

Cited by 5 Pith papers

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

  1. CARD: Cross-component Audio Representation Distillation for Encoder-Free Audio Captioning

    cs.SD 2026-07 conditional novelty 6.0 of 10

    Routing early CLAP stages to the audio projector and later stages to the LLM yields encoder-free captioning that beats LLM-only distillation by +12.2 CIDEr-D on AudioCaps.

  2. Language-Unlocked ViT (LUViT): Empowering Self-Supervised Vision Transformers with LLMs

    cs.CV 2025-07 conditional novelty 6.0 of 10

    LUViT jointly pretrains a ViT with masked auto-encoding and LoRA adapters in a frozen LLM block, reporting +0.4% ImageNet-1K accuracy and up to +2.2% on ImageNet-A over its own MAE baseline.

  3. Mono-InternVL-1.5: Towards Cheaper and Faster Monolithic Multimodal Large Language Models

    cs.CV 2025-07 conditional novelty 5.5 of 10

    A monolithic multimodal LLM that cuts pre-training data by 58% and first-token latency by up to 69% while matching or beating its predecessor on 15 benchmarks.

  4. LPCAN: Lightweight Pyramid Cross-Attention Network for Rail Surface Defect Detection Using RGB-D Data

    cs.CV 2026-01 reject novelty 4.0 of 10

    A lightweight RGB-D cross-attention network is proposed for rail defect detection, but the SOTA accuracy and generalization claims are internally inconsistent and the implementation is not public.

  5. Knowledge-Embedded and Hypernetwork-Guided Few-Shot Substation Meter Defect Image Generation Method

    cs.CV 2026-01 reject novelty 3.0 of 10

    Fine-tuning Stable Diffusion with DreamBooth-style knowledge and hypernetwork-guided crack control maps can synthesize substation meter defect images that boost a YOLOv8 defect detector's mAP when added to the training set.

Pith tools