mlsanityData to 14 Oct 2026

TNG-CLIP:Training-Time Negation Data Generation for Negation Awareness of CLIP

arXiv 2505.18434 ↗First posted 24 May 2025cs.CV cs.AIcited by 7

On the map: Multimodal LLMs / vision-language models → Prompt learning and adaptation of CLIP

Abstract (arXiv)

Vision-language models (VLMs), such as CLIP, have demonstrated strong performance across a range of downstream tasks. However, CLIP is still limited in negation understanding: the ability to recognize the absence or exclusion of a concept. Existing methods address the problem by using a large language model (LLM) to generate large-scale data of image captions containing negation for further fine-tuning CLIP. However, these methods are both time- and compute-intensive, and their evaluations are typically restricted to image-text matching tasks. To expand the horizon, we (1) introduce a training-time negation data generation pipeline such that negation captions are generated during the training stage, which only increases 2.5% extra training time, and (2) we propose the first benchmark, Neg-TtoI, for evaluating text-to-image generation models on prompts containing negation, assessing model's ability to produce semantically accurate images. We show that our proposed method, TNG-CLIP, achieves SOTA performance on diverse negation benchmarks of image-to-text matching, text-to-image retrieval, and image generation.

Key facts. Cited by 7 papers in the map (data through 14 Oct 2026). Placed in Prompt learning and adaptation of CLIP, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 27 of its 46 references that are in the map. First posted on arXiv 24 May 2025.

Builds on

27 of 46 references.

Closest ancestors · 3

Same direction, newest first

  1. 2025Vision-Language Models Do Not Understand Negation16 Jan
  2. 2024Learn "No" to Say "Yes" Better: Improving Vision-Language Models via Negations29 Mar
  3. 2022When and why vision-language models behave like bags-of-words, and what to do about it?4 Oct

Other references · 24

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2020Language Models are Few-Shot Learnerscited by 32,419
  3. 2022Training language models to follow instructions with human feedbackcited by 15,971
  4. 2015Microsoft COCO Captions: Data Collection and Evaluation Servercited by 1,462
  5. 2015Flickr30k Entities: Collecting Region-to-Phrase Correspondences for Richer Image-to-Sentence Modelscited by 1,362
  6. 2019Self-training with Noisy Student improves ImageNet classificationcited by 1,311
  7. 2021Conceptual 12M: Pushing Web-Scale Image-Text Pre-Training To Recognize Long-Tail Visual Conceptscited by 1,216
  8. 2022Fine-Tuning can Distort Pretrained Features and Underperform Out-of-Distributioncited by 739
  9. 2023TIFA: Accurate and Interpretable Text-to-Image Faithfulness Evaluation with Question Answeringcited by 428
  10. 2017Deep Learning is Robust to Massive Label Noisecited by 165
  11. 2021VALSE: A Task-Independent Benchmark for Vision and Language Models Centered on Linguistic Phenomenacited by 135
  12. 2018Dialog-based Interactive Image Retrievalcited by 84
12 more
  1. 2023Language models are not naysayers: An analysis of language models on negation benchmarkscited by 74
  2. 2024Adaptive Data Optimization: Dynamic Sample Selection with Scaling Lawscited by 47
  3. 2025Know "No" Better: A Data-Driven Approach for Enhancing Negation Awareness in CLIPcited by 28
  4. 2022Not another Negation Benchmark: The NaN-NLI Test Suite for Sub-clausal Negationcited by 25
  5. 2024How and where does CLIP process negation?cited by 16
  6. 2024Investigating and Addressing Hallucinations of LLMs in Tasks Involving Negationcited by 15
  7. 2023Beyond Positive Scaling: How Negation Impacts Scaling Trends of Language Modelscited by 12
  8. 2023Task-Attentive Transformer Architecture for Continual Learning of Vision-and-Language Tasks Using Knowledge Distillationcited by 11
  9. 2015WhittleSearch: Interactive Image Search with Relative Attribute Feedbackcited by 10
  10. 2024Impact of Noisy Supervision in Foundation Model Learningcited by 8
  11. 2018Dynamic Sentence Sampling for Efficient Training of Neural Machine Translationcited by 6
  12. 2023Modyn: Data-Centric Machine Learning Pipeline Orchestrationcited by 5

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 7.

7 papers.

  1. 2026DEO: Training-Free Direct Embedding Optimization for Negation-Aware Retrievalcited by 2
  2. 2026NAST: Improving Negation Handling in Medical Vision-Language Models through Negation-Aware Selective Trainingcited by 0
  3. 2026Exploring the AI Obedience: Why is Generating a Pure Color Image Harder than CyberPunk?cited by 0
  4. 2026Benchmarking MLLMs via Cognitive Expected Scene Graph for Safety-Critical Visual Negation Understandingcited by 0
  5. 2026Absence is Presence: Understanding Visual Scene Negative Events Under Safety Cognitive Constraintcited by 0
  6. 2026Skeleton-and-Strategy Prompting: Training-Free Negation Understanding for Vision-Language Modelscited by 0
  7. 2026NegT2IBench: When Negation Changes the Picture. A Polarity Benchmark for Text-to-Image Modelscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026When Negation Is a Geometry Problem in Vision-Language Models
  2. 2026Not Just What's There: Enabling CLIP to Comprehend Negated Visual Descriptions Without Fine-tuning
  3. 2026Omni-NegCLIP: Enhancing CLIP with Front-Layer Contrastive Fine-Tuning for Comprehensive Negation Understanding
  4. 2025SpaceVLM: Sub-Space Modeling of Negation in Vision-Language Models
  5. 2025The Effect of Negation on CLIP in Medical Imaging: Limitations of Contrastive Language-Image Pretraining
  6. 2026HANCLIP: A Family of Hyperbolic Angular Negation Vision Language Models
  7. 2025Contrastive vision-language learning with paraphrasing and negation
  8. 2026CLIP Is Shortsighted: Paying Attention Beyond the First Sentence