# KeySI: An Interaction Framework for Tuning Text Embeddings Based on Human Feedback

arXiv: https://arxiv.org/abs/2607.20556 · HTML: https://mlsanity.com/paper/2607.20556/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Human values and AI feedback alignment, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 9 of its 44 references that are in the map.
- First posted on arXiv 17 Jul 2026.

## Placed in

- LLM alignment / RLHF / preference optimisation / [Human values and AI feedback alignment](https://mlsanity.com/alignment/human-values-and-ai-feedback-alignment/)

## Builds on (most cited first, 9 of 9)

- [Concept Induction: Analyzing Unstructured Text with High-Level Concepts Using LLooM](https://mlsanity.com/paper/2404.12259/) (2024)
- [Semantic Concept Spaces: Guided Topic Model Refinement using Word-Embedding Projections](https://mlsanity.com/paper/1908.00475/) (2019)
- [HEIDL: Learning Linguistic Expressions with Deep Learning and Human-in-the-Loop](https://mlsanity.com/paper/1907.11184/) (2019)
- [DeepSI: Interactive Deep Learning for Semantic Interaction](https://mlsanity.com/paper/2305.18357/) (2023)
- [A Survey of Pre-trained Language Models for Processing Scientific Text](https://mlsanity.com/paper/2401.17824/) (2024)
- [DeepVA: Bridging Cognition and Computation through Semantic Interaction and Deep Learning](https://mlsanity.com/paper/2007.15800/) (2020)
- [ImageSI: Semantic Interaction for Deep Learning Image Projections](https://mlsanity.com/paper/2408.03845/) (2024)
- [Semantic Interactive Learning for Text Classification: A Constructive Approach for Contextual Interactions](https://mlsanity.com/paper/2209.02984/) (2022)
- [NeuralSI: Neural Design of Semantic Interaction for Interactive Deep Learning](https://mlsanity.com/paper/2402.17178/) (2024)

## Abstract (arXiv)

In large-scale text analysis tasks, pre-trained language models are often used to embed text corpora for downstream analysis. However, such models may struggle to capture domain-specific semantics and adapting them typically requires large amounts of labeled data and technical expertise to implement training pipelines. Recent approaches have demonstrated how visual interactions in document projections can capture human feedback as training signals for model tuning. However, these methods operate on document-level feedback, which requires users to open and assess individual documents in order to provide effective feedback. In this paper, we propose KeySI, an interaction framework that enables feature-level feedback through keyword-based concept specification. Users specify feedback by organizing extracted keywords into groups representing concepts, which KeySI translates into document-level supervision for subsequent tuning. By operating on keywords as the primary interaction medium, KeySI reduces the need for manual document inspection and labeling and lowers the barrier to adapting embedding models. We present a prototype implementation that, given a corpus, curates representative keywords, visualizes keywords and document embeddings via dimensionality reduction, allows interactive specification of keyword groups, and supports iterative refinement through system feedback. We evaluate KeySI through a user study, usage scenarios, and quantitative experiments demonstrating its effectiveness in capturing user intent and improving embedding alignment.