# Beyond Demonstrations: Dynamic Vector Construction from Latent Representations

arXiv: https://arxiv.org/abs/2505.20318 · HTML: https://mlsanity.com/paper/2505.20318/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 14 of its 22 references that are in the map.
- First posted on arXiv 23 May 2025.

## Builds on (most cited first, 14 of 14)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Self-Consistency Improves Chain of Thought Reasoning in Language Models](https://mlsanity.com/paper/2203.11171/) (2022)
- [Character-level Convolutional Networks for Text Classification](https://mlsanity.com/paper/1509.01626/) (2015)
- [Inference-Time Intervention: Eliciting Truthful Answers from a Language Model](https://mlsanity.com/paper/2306.03341/) (2023)
- [What Makes Good In-Context Examples for GPT-3?](https://mlsanity.com/paper/2101.06804/) (2021)
- [Steering Llama 2 via Contrastive Activation Addition](https://mlsanity.com/paper/2312.06681/) (2023)
- [Few-Shot Parameter-Efficient Fine-Tuning is Better and Cheaper than In-Context Learning](https://mlsanity.com/paper/2205.05638/) (2022)
- [Learning To Retrieve Prompts for In-Context Learning](https://mlsanity.com/paper/2112.08633/) (2021)
- [In-Context Learning Creates Task Vectors](https://mlsanity.com/paper/2310.15916/) (2023)
- [In-context Vectors: Making In Context Learning More Effective and Controllable Through Latent Space Steering](https://mlsanity.com/paper/2311.06668/) (2023)
- [Steering Away from Harm: An Adaptive Approach to Defending Vision Language Model Against Jailbreaks](https://mlsanity.com/paper/2411.16721/) (2024)
- [Designing Role Vectors to Improve LLM Inference Behaviour](https://mlsanity.com/paper/2502.12055/) (2025)

## Built on by (most cited first, 4 of 4)

- [Task Vectors, Learned Not Extracted: Performance Gains and Mechanistic Insight](https://mlsanity.com/paper/2509.24169/) (2025)
- [RepBench: Compiling Benchmarks into Capability Representations for Large Language Models](https://mlsanity.com/paper/2607.28008/) (2026)
- [Efficient Task Adaptation in Large Language Models: A Survey of Weight-Based, Prompt-Based, and Embedding-Based Adaptations](https://mlsanity.com/paper/2610.00928/) (2026)
- [Capturing In-Context Learning Dynamics with Task Operators](https://mlsanity.com/paper/2610.01054/) (2026)

## Abstract (arXiv)

In-Context derived Vector (ICV) methods extract task-relevant representations from large language models (LLMs) and reinject them during inference, achieving comparable performance to few-shot In-Context Learning (ICL) without repeated demonstration processing. However, existing ICV methods remain sensitive to ICL-specific factors, often use coarse or semantically fragmented representations as the source of the vector, and rely on heuristic-based injection positions, limiting their applicability. To address these issues, we propose Dynamic Vector (DyVec), which incorporates an Exhaustive Query Rotation (EQR) strategy to extract robust semantically aggregated latent representations by mitigating variance introduced by ICL. It then applies Dynamic Latent Segmentation and Injection to adaptively partition representations based on task complexity and leverages REINFORCE-based optimization to learn optimal injection positions for each segment. Experiments results show that DyVec outperforms few-shot ICL, LoRA, and prior ICV baselines. Further analysis highlights the effectiveness of dynamically segmenting and injecting semantically aggregated latent representations. DyVec provides a lightweight and data-efficient solution for inference-time task adaptation.