Pith. sign in

REVIEW 16 cited by

Long-CLIP: Unlocking the Long-Text Capability of CLIP

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 2403.15378 v3 pith:E7PSPX7P submitted 2024-03-22 cs.CV

Long-CLIP: Unlocking the Long-Text Capability of CLIP

classification cs.CV
keywords cliplong-cliptexttext-imageretrievallengthsignificantcapabilities
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Contrastive Language-Image Pre-training (CLIP) has been the cornerstone for zero-shot classification, text-image retrieval, and text-image generation by aligning image and text modalities. Despite its widespread adoption, a significant limitation of CLIP lies in the inadequate length of text input. The length of the text token is restricted to 77, and an empirical study shows the actual effective length is even less than 20. This prevents CLIP from handling detailed descriptions, limiting its applications for image retrieval and text-to-image generation with extensive prerequisites. To this end, we propose Long-CLIP as a plug-and-play alternative to CLIP that supports long-text input, retains or even surpasses its zero-shot generalizability, and aligns the CLIP latent space, making it readily replace CLIP without any further adaptation in downstream frameworks. Nevertheless, achieving this goal is far from straightforward, as simplistic fine-tuning can result in a significant degradation of CLIP's performance. Moreover, substituting the text encoder with a language model supporting longer contexts necessitates pretraining with vast amounts of data, incurring significant expenses. Accordingly, Long-CLIP introduces an efficient fine-tuning solution on CLIP with two novel strategies designed to maintain the original capabilities, including (1) a knowledge-preserved stretching of positional embedding and (2) a primary component matching of CLIP features. With leveraging just one million extra long text-image pairs, Long-CLIP has shown the superiority to CLIP for about 20% in long caption text-image retrieval and 6% in traditional text-image retrieval tasks, e.g., COCO and Flickr30k. Furthermore, Long-CLIP offers enhanced capabilities for generating images from detailed text descriptions by replacing CLIP in a plug-and-play manner.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 16 Pith papers

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

  1. VideoNet: A Large-Scale Dataset for Domain-Specific Action Recognition

    cs.CV 2026-05 unverdicted novelty 7.0

    VideoNet is a new large-scale benchmark and training dataset for domain-specific action recognition that exposes limitations in VLMs and enables smaller fine-tuned models to surpass larger open-weight ones.

  2. LFS: Learnable Frame Selector for Event-Aware and Temporally Diverse Video Captioning

    cs.CV 2026-01 conditional novelty 7.0

    LFS learns to select temporally diverse and event-aware frames for video captioning by using direct feedback from frozen video-LLMs, yielding gains up to 2% on VDC and over 4% on the new ICH-CC benchmark.

  3. HPP: Hierarchical Programmatic Probing for Long Video Understanding by Decoupling Perception and Reasoning

    cs.CV 2026-06 unverdicted novelty 6.0

    HPP decouples perception from reasoning in long-video VLMs by having an LLM run iterative programmatic probes on hierarchically segmented video, reporting gains on LongVideoBench, EgoSchema, VideoMME, and MLVU.

  4. Offline Semantic Guidance for Efficient Vision-Language-Action Policy Distillation

    cs.CV 2026-05 conditional novelty 6.0

    VLA-AD distills 7B VLA teachers into 158M students using offline VLM semantic guidance on task phases and directions, matching teacher performance on LIBERO with 44x size reduction and 3.28x speedup.

  5. Delta Forcing: Trust Region Steering for Interactive Autoregressive Video Generation

    cs.CV 2026-05 unverdicted novelty 6.0

    Delta Forcing uses latent trajectory deltas to adaptively limit unreliable teacher guidance while enforcing monotonic continuity, improving temporal consistency in interactive autoregressive video generation.

  6. Delta Forcing: Trust Region Steering for Interactive Autoregressive Video Generation

    cs.CV 2026-05 unverdicted novelty 6.0

    Delta Forcing estimates transition consistency from the latent delta between teacher and generator trajectories to balance teacher supervision against a monotonic continuity objective in autoregressive video generation.

  7. Delta Forcing: Trust Region Steering for Interactive Autoregressive Video Generation

    cs.CV 2026-05 unverdicted novelty 6.0

    Delta Forcing improves temporal coherence in interactive autoregressive video generation by estimating transition consistency from teacher-generator latent deltas and balancing it against a monotonic continuity objective.

  8. AnyStyle: Single-Pass Multimodal Stylization for 3D Gaussian Splatting

    cs.CV 2026-02 conditional novelty 6.0

    A single-pass 3D Gaussian splatting pipeline that stylizes unposed scenes from either a text prompt or a reference image via a lightweight zero-initialized style-injection branch.

  9. Zero-Effort Image-to-Music Generation: An Interpretable RAG-based VLM Approach

    cs.SD 2025-09 unverdicted novelty 6.0

    A zero-training VLM framework generates music from images via ABC notation, multi-modal RAG, and self-refinement while providing text and visual explanations for the outputs.

  10. E5-V: Universal Embeddings with Multimodal Large Language Models

    cs.CL 2024-07 unverdicted novelty 6.0

    E5-V produces strong universal multimodal embeddings from MLLMs trained solely on text pairs, often surpassing prior methods across retrieval and related tasks without multimodal fine-tuning.

  11. JuZhou 1.0 Technical Report: The First Edge-Native Text-to-Image Foundation Model Trained Entirely on China-Developed AI Accelerators

    cs.CV 2026-06 conditional novelty 5.0

    A compact Chinese-native T2I U-Net (~0.387B) trained on Sugon K100, distilled to 4 steps, reports GenEval 0.69 and ~1.6–4.5s offline mobile generation.

  12. TuringViT: Making SOTA Vision Transformers Accessible to All

    cs.CV 2026-06 unverdicted novelty 5.0

    TuringViT uses Turing Linear Attention, VISTA-Curation, and dynamic-resolution pretraining to outperform open ViT baselines with 10% data while improving VLM performance and high-resolution latency.

  13. TuringViT: Making SOTA Vision Transformers Accessible to All

    cs.CV 2026-06 unverdicted novelty 5.0

    TuringViT claims a new ViT design with linear attention and curated data that matches SOTA performance using 10% of typical pretraining data while supporting dynamic resolutions and improving VLM integration.

  14. InternLM-XComposer-2.5: A Versatile Large Vision Language Model Supporting Long-Contextual Input and Output

    cs.CV 2024-07 conditional novelty 5.0

    InternLM-XComposer-2.5 is a 7B vision-language model supporting up to 96K context that reaches GPT-4V-level performance on image, video, and multi-turn tasks and adds LoRA-driven text-image composition capabilities.

  15. JuZhou 1.0 Technical Report: The First Edge-Native Text-to-Image Foundation Model Trained Entirely on China-Developed AI Accelerators

    cs.CV 2026-06 unverdicted novelty 4.0

    JuZhou 1.0 is a 0.387B-parameter T2I diffusion model with 4-step inference achieving 0.69 GenEval, trained on 9M Chinese pairs using Sugon K100 accelerators and deployable on Android/iOS devices.

  16. Delta Forcing: Trust Region Steering for Interactive Autoregressive Video Generation

    cs.CV 2026-05 unverdicted novelty 4.0

    Delta Forcing constrains unreliable teacher supervision in autoregressive video models using an adaptive trust region based on latent trajectory deltas to improve consistency without losing event responsiveness.