# Head Pursuit: Probing Attention Specialization in Multimodal Transformers

arXiv: https://arxiv.org/abs/2510.21518 · HTML: https://mlsanity.com/paper/2510.21518/

## Key facts

- Cited by 17 papers in the map (data through 14 Oct 2026).
- Placed in MLLM training and visual instruction tuning, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 36 of its 51 references that are in the map.
- First posted on arXiv 24 Oct 2025.

## Placed in

- Multimodal LLMs / vision-language models / [MLLM training and visual instruction tuning](https://mlsanity.com/mllm/mllm-training-and-visual-instruction-tuning/)

## Builds on (most cited first, 30 of 36)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [Flamingo: a Visual Language Model for Few-Shot Learning](https://mlsanity.com/paper/2204.14198/) (2022)
- [TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehension](https://mlsanity.com/paper/1705.03551/) (2017)
- [Locating and Editing Factual Associations in GPT](https://mlsanity.com/paper/2202.05262/) (2022)
- [Describing Textures in the Wild](https://mlsanity.com/paper/1311.3618/) (2013)
- [CIDEr: Consensus-based Image Description Evaluation](https://mlsanity.com/paper/1411.5726/) (2014)
- [Gemma 3 Technical Report](https://mlsanity.com/paper/2503.19786/) (2025)
- [EuroSAT: A Novel Dataset and Deep Learning Benchmark for Land Use and Land Cover Classification](https://mlsanity.com/paper/1709.00029/) (2017)
- [RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Models](https://mlsanity.com/paper/2009.11462/) (2020)
- [Analyzing Multi-Head Self-Attention: Specialized Heads Do the Heavy Lifting, the Rest Can Be Pruned](https://mlsanity.com/paper/1905.09418/) (2019)
- [Are Sixteen Heads Really Better than One?](https://mlsanity.com/paper/1905.10650/) (2019)
- [Remote Sensing Image Scene Classification: Benchmark and State of the Art](https://mlsanity.com/paper/1703.00121/) (2017)
- [The Linear Representation Hypothesis and the Geometry of Large Language Models](https://mlsanity.com/paper/2311.03658/) (2023)
- [Eliciting Latent Predictions from Transformers with the Tuned Lens](https://mlsanity.com/paper/2303.08112/) (2023)
- [Dissecting Recall of Factual Associations in Auto-Regressive Language Models](https://mlsanity.com/paper/2304.14767/) (2023)
- [Interpreting CLIP's Image Representation via Text-Based Decomposition](https://mlsanity.com/paper/2310.05916/) (2023)
- [CRAFT: Concept Recursive Activation FacTorization for Explainability](https://mlsanity.com/paper/2211.10154/) (2022)
- [Analyzing Transformers in Embedding Space](https://mlsanity.com/paper/2209.02535/) (2022)
- [Towards Interpreting Visual Information Processing in Vision-Language Models](https://mlsanity.com/paper/2410.07149/) (2024)
- [On the Origins of Linear Representations in Large Language Models](https://mlsanity.com/paper/2403.03867/) (2024)
- [Understanding Information Storage and Transfer in Multi-modal Large Language Models](https://mlsanity.com/paper/2406.04236/) (2024)
- [Talking Heads: Understanding Inter-layer Communication in Transformer Language Models](https://mlsanity.com/paper/2406.09519/) (2024)
- [Competition of Mechanisms: Tracing How Language Models Handle Facts and Counterfactuals](https://mlsanity.com/paper/2402.11655/) (2024)
- [A Concept-Based Explainability Framework for Large Multimodal Models](https://mlsanity.com/paper/2406.08074/) (2024)
- [Interpreting the linear structure of vision-language model embedding spaces](https://mlsanity.com/paper/2504.11695/) (2025)
- [Attention Lens: A Tool for Mechanistically Interpreting the Attention Head Information Retrieval Mechanism](https://mlsanity.com/paper/2310.16270/) (2023)
- [Analyzing Finetuning Representation Shift for Multimodal LLMs Steering](https://mlsanity.com/paper/2501.03012/) (2025)

## Built on by (most cited first, 17 of 17)

- [Large Vision-Language Models Get Lost in Attention](https://mlsanity.com/paper/2605.05668/) (2026)
- [Towards Understanding Modality Interaction in Multimodal Language Models via Partial Information Decomposition](https://mlsanity.com/paper/2606.00959/) (2026)
- [Are Audio-Language Models Listening? Audio-Specialist Heads for Adaptive Audio Steering](https://mlsanity.com/paper/2603.06854/) (2026)
- [Mechanistic Insights into Functional Sparsity in Multimodal LLMs via CoRe Heads](https://mlsanity.com/paper/2606.05843/) (2026)
- [Multimodal In-context Learning for ASR of Low-resource Languages](https://mlsanity.com/paper/2601.05707/) (2026)
- [Interpreting Transformers Through Attention Head Intervention](https://mlsanity.com/paper/2601.04398/) (2026)
- [Softmax Linear Attention: Reclaiming Global Competition](https://mlsanity.com/paper/2602.01744/) (2026)
- [Embedding-Aware Feature Discovery: Bridging Latent Representations and Interpretable Features in Event Sequences](https://mlsanity.com/paper/2603.15713/) (2026)
- [Knowing Isn't Always Saying: When Do Spatial Encodings Reach Answers in Vision-Language Models?](https://mlsanity.com/paper/2608.22916/) (2026)
- [Semantic Head Specialization Guides Hybrid ViT Attention for Multimodal LLMs](https://mlsanity.com/paper/2608.28383/) (2026)
- [CALM: Class-Conditional Sparse Attention Vectors for Large Audio-Language Models](https://mlsanity.com/paper/2602.07077/) (2026)
- [Do Foundation Models Know Geometry? Probing Frozen Features for Continuous Physical Measurement](https://mlsanity.com/paper/2603.06459/) (2026)
- [HaloProbe: Bayesian Detection and Mitigation of Object Hallucinations in Vision-Language Models](https://mlsanity.com/paper/2604.06165/) (2026)
- [Mitigating Action-Relation Hallucinations in LVLMs via Relation-aware Visual Enhancement](https://mlsanity.com/paper/2605.11808/) (2026)
- [LocUS: Head Selection and Subspace Projection for Targeted Activation Steering](https://mlsanity.com/paper/2609.31122/) (2026)
- [Look Before You Judge: Training-Free Region Mining for Grounded and Explainable Deepfake Detection](https://mlsanity.com/paper/2609.35536/) (2026)
- [Emergent Structure in the Marginal Attention Space of Language Models](https://mlsanity.com/paper/2610.03109/) (2026)

## Abstract (arXiv)

Language and vision-language models have shown impressive performance across a wide range of tasks, but their internal mechanisms remain only partly understood. In this work, we study how individual attention heads in text-generative models specialize in specific semantic or visual attributes. Building on an established interpretability method, we reinterpret the practice of probing intermediate activations with the final decoding layer through the lens of signal processing. This lets us analyze multiple samples in a principled way and rank attention heads based on their relevance to target concepts. Our results show consistent patterns of specialization at the head level across both unimodal and multimodal transformers. Remarkably, we find that editing as few as 1% of the heads, selected using our method, can reliably suppress or enhance targeted concepts in the model output. We validate our approach on language tasks such as question answering and toxicity mitigation, as well as vision-language tasks including image classification and captioning. Our findings highlight an interpretable and controllable structure within attention layers, offering simple tools for understanding and editing large-scale generative models.