Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Videos

Negation-Aware Test-Time Adaptation for Vision-Language Models.

Haochen Han, Alex Jinpeng Wang, Fangming Liu

IEEE Transactions on Pattern Analysis and Machine Intelligence

|June 12, 2026

Summary

This summary is machine-generated.

Related Concept Videos

Improving Translational Accuracy

Improving Translational Accuracy

Base complementarity between the three base pairs of mRNA codon and the tRNA anticodon is not a failsafe mechanism. Inaccuracies can range from a single mismatch to no correct base pairing at all. The free energy difference between the correct and nearly correct base pairs can be as small as 3 kcal/ mol. With complementarity being the only proofreading step, the estimated error frequency would be one wrong amino acid in every 100 amino acids incorporated. However, error frequencies observed in...

Improving Translational Accuracy

Improving Translational Accuracy

Base complementarity between the three base pairs of mRNA codon and the tRNA anticodon is not a failsafe mechanism. Inaccuracies can range from a single mismatch to no correct base pairing at all. The free energy difference between the correct and nearly correct base pairs can be as small as 3 kcal/ mol. With complementarity being the only proofreading step, the estimated error frequency would be one wrong amino acid in every 100 amino acids incorporated. However, error frequencies observed in...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

A Retrospective Follow-up Study on the Effects of Different Intravesical Treatments on Recurrence Control and Quality of Life in Patients With Non-Muscle-Invasive Bladder Cancer After Transurethral Resection of Bladder Tumour.

Archivos espanoles de urologia·2026

Same author

Single-cell identifies and validates human circulating Treg subtype/state Treg<sup>fci</sup> in non-small cell lung cancer.

Signal transduction and targeted therapy·2026

Same author

Ferroptosis in musculoskeletal disorders: Emerging mechanisms and therapeutic opportunities (Review).

International journal of molecular medicine·2026

Same author

Stage-Resolved Metabolomics of Fruit Development and Oil Accumulation in Idesia polycarpa.

Physiologia plantarum·2026

Same author

Effects of different land use types on soil microorganisms and physicochemical properties in the Huaihe River source region.

BMC microbiology·2026

Same author

Polydopamine Nanocages Orchestrate Multi-Site Networks on the Ni<sub>4</sub>Mo Electrocatalyst for Efficient and Ultrastable Hydrogen Evolution.

Journal of the American Chemical Society·2026

Same journal

TraGraph-GS: Trajectory Graph-based Gaussian Splatting for Arbitrary Large-Scale Scene Rendering.

IEEE transactions on pattern analysis and machine intelligence·2026

Same journal

SWIFT: A Small-World Interaction Framework for Flow-Aware Trajectory Prediction in Autonomous Driving.

IEEE transactions on pattern analysis and machine intelligence·2026

Same journal

HardFlow: Hard-Constrained Sampling for Flow-Matching Models Via Trajectory Optimization.

IEEE transactions on pattern analysis and machine intelligence·2026

Same journal

Industrial Brain: Self-Evolving Neuro-Symbolic Autonomy with Causal Resilience for Cyber-Physical Systems.

IEEE transactions on pattern analysis and machine intelligence·2026

Same journal

Adaptive Hardness-Driven Dictionary Distillation for Incomplete Streaming View Clustering.

IEEE transactions on pattern analysis and machine intelligence·2026

Same journal

Mixture of Global and Local Experts with Diffusion Transformer for Controllable Face Generation.

IEEE transactions on pattern analysis and machine intelligence·2026

See all related articles

This study introduces Negation-Aware Test-Time Adaptation (NEAT) to improve vision-language models' negation understanding. NEAT efficiently adjusts parameters during inference, enhancing performance with minimal computational cost.

Area of Science:

Computer Science
Artificial Intelligence
Machine Learning

Background:

Vision-Language Models (VLMs) struggle with negation understanding, limiting real-world applications.
Existing solutions require extensive data and computational resources for fine-tuning.
The core challenge lies in dual-concept shifts between affirmation and negation distributions.

Purpose of the Study:

To develop an efficient method for improving negation understanding in VLMs.
To address the limitations of data-centric approaches for negation handling.
To propose a low-carbon solution for negation understanding in AI models.

Main Methods:

Proposed Negation-Aware Test-Time Adaptation (NEAT) method.
Adjusts distribution-related parameters during inference to handle negation.

Related Experiment Videos

Reduces distribution shifts in consistent semantics while eliminating false distributional consistency.

Main Results:

NEAT demonstrates effectiveness on various negation understanding tasks.
Achieves comparable or superior performance to state-of-the-art post-training methods.
Requires less than 0.01% of trainable parameters, indicating high efficiency.

Conclusions:

NEAT offers an efficient and effective solution for VLM negation understanding.
The method addresses the dual-concept shift problem with minimal resource requirements.
NEAT represents a significant advancement in low-carbon AI for practical applications.