# When More Words Say Less: Decoupling Length and Specificity in Image Description Evaluation

arXiv: https://arxiv.org/abs/2601.04609 · HTML: https://mlsanity.com/paper/2601.04609/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in VLM benchmarks and visual reasoning, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 22 of its 62 references that are in the map.
- First posted on arXiv 8 Jan 2026.

## Placed in

- Multimodal LLMs / vision-language models / [VLM benchmarks and visual reasoning](https://mlsanity.com/mllm/vlm-benchmarks-and-visual-reasoning/)

## Builds on (most cited first, 22 of 22)

- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution](https://mlsanity.com/paper/2409.12191/) (2024)
- [LAION-5B: An open large-scale dataset for training next generation image-text models](https://mlsanity.com/paper/2210.08402/) (2022)
- [CLIPScore: A Reference-free Evaluation Metric for Image Captioning](https://mlsanity.com/paper/2104.08718/) (2021)
- [Molmo and PixMo: Open Weights and Open Data for State-of-the-Art Vision-Language Models](https://mlsanity.com/paper/2409.17146/) (2024)
- [A Long Way to Go: Investigating Length Correlations in RLHF](https://mlsanity.com/paper/2310.03716/) (2023)
- [Concise Thoughts: Impact of Output Length on LLM Reasoning and Cost](https://mlsanity.com/paper/2407.19825/) (2024)
- [Loose lips sink ships: Mitigating Length Bias in Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2310.05199/) (2023)
- [Reasoning About Pragmatics with Neural Listeners and Speakers](https://mlsanity.com/paper/1604.00562/) (2016)
- [Explaining Length Bias in LLM-Based Preference Evaluations](https://mlsanity.com/paper/2407.01085/) (2024)
- [ALOHa: A New Measure for Hallucination in Captioning Models](https://mlsanity.com/paper/2404.02904/) (2024)
- [Deep Learning Approaches on Image Captioning: A Review](https://mlsanity.com/paper/2201.12944/) (2022)
- [Pragmatically Informative Image Captioning with Character-Level Inference](https://mlsanity.com/paper/1804.05417/) (2018)
- [When redundancy is useful: A Bayesian approach to 'overinformative' referring expressions](https://mlsanity.com/paper/1903.08237/) (2019)
- [Context Matters for Image Descriptions for Accessibility: Challenges for Referenceless Evaluation Metrics](https://mlsanity.com/paper/2205.10646/) (2022)
- [Grow-and-Clip: Informative-yet-Concise Evidence Distillation for Answer Explanation](https://mlsanity.com/paper/2201.05088/) (2022)
- [Pragmatic Issue-Sensitive Image Captioning](https://mlsanity.com/paper/2004.14451/) (2020)
- [Grounding Language in Multi-Perspective Referential Communication](https://mlsanity.com/paper/2410.03959/) (2024)
- [An Examination of the Robustness of Reference-Free Image Captioning Evaluation Metrics](https://mlsanity.com/paper/2305.14998/) (2023)
- [Communication-based Evaluation for Natural Language Generation](https://mlsanity.com/paper/1909.07290/) (2019)
- [Harnessing Caption Detailness for Data-Efficient Text-to-Image Generation](https://mlsanity.com/paper/2505.15172/) (2025)
- [Do Vision Encoders Truly Explain Object Hallucination?: Mitigating Object Hallucination via Simple Fine-Grained CLIPScore](https://mlsanity.com/paper/2502.20034/) (2025)

## Built on by (most cited first, 2 of 2)

- [PROMPT2BOX:Improving LLM Weakness Discovery and Specificity Estimation by Uncovering Entailment Structure among Prompts](https://mlsanity.com/paper/2603.21438/) (2026)
- [Learning to Refer from Estimated Listener Gaze](https://mlsanity.com/paper/2609.14207/) (2026)

## Abstract (arXiv)

Vision-language models (VLMs) are increasingly used to make visual content accessible via text-based descriptions. In current systems, however, description specificity is often conflated with their length. We argue that these two concepts must be disentangled: descriptions can be concise yet dense with information, or lengthy yet vacuous. We define specificity relative to a contrast set, where a description is more specific to the extent that it picks out the target image better than other possible images. We construct a dataset that controls for length while varying information content, and validate that people reliably prefer more specific descriptions regardless of length. We find that controlling for length alone cannot account for differences in specificity: how the length budget is allocated makes a difference. These results support evaluation approaches that directly prioritize specificity over verbosity.