mlsanityData to 14 Oct 2026

An Empirical Study on Configuring In-Context Learning Demonstrations for Unleashing MLLMs' Sentimental Perception Capability

arXiv 2505.16193 ↗First posted 22 May 2025cs.CL cs.CVcited by 4

On the map: Multimodal LLMs / vision-language models → Multimodal emotion understanding

Abstract (arXiv)

The advancements in Multimodal Large Language Models (MLLMs) have enabled various multimodal tasks to be addressed under a zero-shot paradigm. This paradigm sidesteps the cost of model fine-tuning, emerging as a dominant trend in practical application. Nevertheless, Multimodal Sentiment Analysis (MSA), a pivotal challenge in the quest for general artificial intelligence, fails to accommodate this convenience. The zero-shot paradigm exhibits undesirable performance on MSA, casting doubt on whether MLLMs can perceive sentiments as competent as supervised models. By extending the zero-shot paradigm to In-Context Learning (ICL) and conducting an in-depth study on configuring demonstrations, we validate that MLLMs indeed possess such capability. Specifically, three key factors that cover demonstrations' retrieval, presentation, and distribution are comprehensively investigated and optimized. A sentimental predictive bias inherent in MLLMs is also discovered and later effectively counteracted. By complementing each other, the devised strategies for three factors result in average accuracy improvements of 15.9% on six MSA datasets against the zero-shot paradigm and 11.2% against the random ICL baseline.

Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Placed in Multimodal emotion understanding, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 38 of its 64 references that are in the map. First posted on arXiv 22 May 2025.

Builds on

38 of 64 references.

References · 38

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2020Language Models are Few-Shot Learnerscited by 32,419
  3. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  4. 2023BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Modelscited by 6,137
  5. 2022Flamingo: a Visual Language Model for Few-Shot Learningcited by 4,386
  6. 2023PaLM-E: An Embodied Multimodal Language Modelcited by 2,071
  7. 2015Show, Attend and Tell: Neural Image Caption Generation with Visual Attentioncited by 1,423
  8. 2021What Makes Good In-Context Examples for GPT-3?cited by 1,195
  9. 2021Fantastically Ordered Prompts and Where to Find Them: Overcoming Few-Shot Prompt Order Sensitivitycited by 1,135
  10. 2023Large Language Modelscited by 780
  11. 2023Language Is Not All You Need: Aligning Perception with Language Modelscited by 543
  12. 2017Tensor Fusion Network for Multimodal Sentiment Analysiscited by 433
26 more
  1. 2023OBELICS: An Open Web-Scale Filtered Dataset of Interleaved Image-Text Documentscited by 316
  2. 2021An Empirical Study of GPT-3 for Few-Shot Knowledge-Based VQAcited by 301
  3. 2022Active Example Selection for In-Context Learningcited by 196
  4. 2022Self-Adaptive In-Context Learning: An Information Compression Perspective for In-Context Example Selection and Orderingcited by 161
  5. 2021What Makes Multi-modal Learning Better than Single (Provably)cited by 153
  6. 2023What In-Context Learning "Learns" In-Context: Disentangling Task Recognition and Task Learningcited by 152
  7. 2018Deep Learning for Sentiment Analysis : A Surveycited by 139
  8. 2022Diverse Demonstrations Improve In-context Compositional Generalizationcited by 139
  9. 2024Exploring the Reasoning Abilities of Multimodal Large Language Models (MLLMs): A Comprehensive Survey on Emerging Trends in Multimodal Reasoningcited by 136
  10. 2023Finding Support Examples for In-Context Learningcited by 104
  11. 2022Z-ICL: Zero-Shot In-Context Learning with Pseudo-Demonstrationscited by 70
  12. 2024What Makes Multimodal In-Context Learning Work?cited by 56
  13. 2021Affective Image Content Analysis: Two Decades Review and New Perspectivescited by 47
  14. 2024EmoVIT: Revolutionizing Emotion Insights with Visual Instruction Tuningcited by 43
  15. 2023GPT-4V with Emotion: A Zero-shot Benchmark for Generalized Emotion Recognitioncited by 38
  16. 2023Revisiting Multimodal Representation in Contrastive Learning: From Patch and Token Embeddings to Finite Discrete Tokenscited by 35
  17. 2023How to Configure Good In-Context Sequence for Visual Question Answeringcited by 33
  18. 2023Learning Emotion Representations from Verbal and Nonverbal Communicationcited by 31
  19. 2024What Factors Affect Multi-Modal In-Context Learning? An In-Depth Explorationcited by 31
  20. 2023Beyond Task Performance: Evaluating and Reducing the Flaws of Large Multimodal Models with In-Context Learningcited by 30
  21. 2022CLMLF:A Contrastive Learning and Multi-Layer Fusion Method for Multimodal Sentiment Detectioncited by 21
  22. 2022Vision-Language Pre-Training for Multimodal Aspect-Based Sentiment Analysiscited by 21
  23. 2022Few-shot Multimodal Sentiment Analysis based on Multimodal Probabilistic Fusion Promptscited by 9
  24. 2023AoM: Detecting Aspect-oriented Information for Multimodal Aspect-Based Sentiment Analysiscited by 9
  25. 2023Manipulating the Label Space for In-Context Classificationcited by 5
  26. 2023A Novel Energy based Model Mechanism for Multi-modal Aspect-Based Sentiment Analysiscited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 2; 2026: 2.

4 papers.

  1. 2025EmoCaliber: Advancing Reliable Visual Emotion Comprehension via Confidence Verbalization and Calibrationcited by 4
  2. 2025Customizing Visual Emotion Evaluation for MLLMs: An Open-vocabulary, Multifaceted, and Scalable Approachcited by 3
  3. 2026Why Multimodal In-Context Learning Lags Behind? Unveiling the Inner Mechanisms and Bottleneckscited by 2
  4. 2026MVEI & EmObserver: Empowering MLLM-Oriented Visual Emotional Intelligence via Emotion Statement Judgementcited by 1

Similar, not cited

Close in wording, no citation either way

  1. 2026MAF: Multimodal Adaptive Few-shot Prompting for Sentiment Analysis with MLLMs
  2. 2024From Introspection to Best Practices: Principled Analysis of Demonstrations in Multimodal In-Context Learning
  3. 2023Towards Understanding In-Context Learning with Contrastive Demonstrations and Saliency Maps
  4. 2024SymDPO: Boosting In-Context Learning of Large Multimodal Models with Symbol Demonstration Direct Preference Optimization
  5. 2023Can Multimodal Large Language Models Truly Perform Multimodal In-Context Learning?
  6. 2026Multimodal Large Language Models as Synthetic Participants in Video-Based Studies: An Evaluation
  7. 2025Multimodal LLMs See Sentiment
  8. 2024AIM: Let Any Multi-modal Large Language Models Embrace Efficient In-Context Learning