Prompt Sensitivity

Follow this topic Ask about this topic

Momentum

9 papers in the last four weeks, down 10% on the four weeks before. 0.1% of all new papers.

Jul 13Week of Sep 28

Latest papers 68

All topics

Prompt Sensitivity in Vision-Language Grounding: How Small Changes in Wording Affect Object Detection
Apr 18, 2026Dawar Jyoti Deka, Amit Sethi, Syed Mohammad AliOpen-Vocabulary Object Detection VLM Robustness
Department of Aerospace Engineering, Indian Institute of Technology Bombay, Mumbai, India · Department of Electrical Engineering, Indian Institute of Technology Bombay, Mumbai, India
Beyond Prompts: Unconditional 3D Inversion for Out-of-Distribution Shapes
Apr 16, 2026Victoria Yue Chen, Emery Pierson, Léopold Maillard +1Prompt Sensitivity 3D Shape Editing
ETH Zürich · 2 École Polytechnique · Dassault Systèmes
Prompt Optimization Is a Coin Flip: Diagnosing When It Helps in Compound AI Systems
Apr 16, 2026Xing Zhang, Guanghui Wang, Yanwei Cui +4Prompt Sensitivity Large Language Model-Guided Optimization
AWS Generative AI Innovation Center · HSBC Holdings Plc., HSBC Technology Center, China.
How Utilitarian Are OpenAI's Models Really? Replicating and Reinterpreting Pfeffer, Krügel, and Uhl (2025)
Mar 24, 2026Johannes HimmelreichPrompt Sensitivity Moral Reasoning in Language Models
Language Model Maps for Prompt-Response Distributions via Log-Likelihood Vectors
Mar 19, 2026Momose Oyama, Yusuke Takase, Hidetoshi ShimodairaPrompt Sensitivity
Kyoto University · RIKEN
Unexplored flaws in multiple-choice VQA make benchmarking unreliable
Nov 27, 2025Fabio Rosenthal, Sebastian Schmidt, Thorsten Graf +3Visual Question Answering Prompt Sensitivity
Technical University of Munich · Volkswagen AG · Helmholtz AI +1
Measuring Intent Comprehension in LLMs
Jun 19, 2025Nadav Kunievsky, James A. EvansPrompt Sensitivity LLM Evaluation
Knowlesdge Lab, University of Chicago, Chicago, Illinois, USA · Knowledge Lab, Data Science Institute, Department of Sociology, University of Chicago, Chicago, Illinois, USA.
Algorithmic Collusion by Large Language Models
Mar 31, 2024Sara Fish, Yannai A. Gonczarowski, Ran I. ShorrerPrompt Sensitivity Algorithmic Collusion
†School of Engineering and Applied Sciences, Harvard University · ‡Department of Economics and Department of Computer Science, Harvard University · §Department of Economics, Penn State University

Apr 18, 2026, cs.CVJ/K move · Enter open · S save

Prompt Sensitivity in Vision-Language Grounding: How Small Changes in Wording Affect Object Detection

Dawar Jyoti Deka, Amit Sethi, Syed Mohammad Ali

Department of Aerospace Engineering, Indian Institute of Technology Bombay, Mumbai, India · Department of Electrical Engineering, Indian Institute of Technology Bombay, Mumbai, India

Open-Vocabulary Object Detection VLM Robustness Prompt Sensitivity Vision-Language Grounding

Open paper AskPDF arXiv

Vision-language models enable open-vocabulary object grounding through natural language queries, under the implicit assumption that semantically equivalent descriptions yield consistent outputs. We examine this assumption using a controlled pipeline combining DETR for object proposals with CLIP for language-conditioned selection on 263 COCO val2017 images. We find that overlapping prompts such as "a person," "a human," and "a pedestrian" frequently select different instances, with mean instability of 2.11 distinct selections across six prompts. PCA analysis shows this variability is structured and directional, not random. Prompt ensembling does not improve quality and often shifts selections toward generic regions. We further show that text embedding proximity explains only 34% of grounding disagreement (r = -0.58), confirming that instability arises from the argmax selection mechanism rather than text-level distances alone.

Momentum

Related topics

Latest papers 68