# Unsupervised Feature Transformation via In-context Generation, Generator-critic LLM Agents, and Duet-play Teaming

arXiv: https://arxiv.org/abs/2504.21304 · HTML: https://mlsanity.com/paper/2504.21304/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- Placed in Web, search, and research agents, one of 16 directions in LLM agents and tool use.
- Builds on 21 of its 28 references that are in the map.
- First posted on arXiv 30 Apr 2025.

## Placed in

- LLM agents and tool use / [Web, search, and research agents](https://mlsanity.com/agents/web-search-and-research-agents/)

## Builds on (most cited first, 21 of 21)

- [RLHF-V: Towards Trustworthy MLLMs via Behavior Alignment from Fine-grained Correctional Human Feedback](https://mlsanity.com/paper/2312.00849/) (2023)
- [Large Language Models in Finance: A Survey](https://mlsanity.com/paper/2311.10723/) (2023)
- [Large Language Models for Automated Data Science: Introducing CAAFE for Context-Aware Automated Feature Engineering](https://mlsanity.com/paper/2305.03403/) (2023)
- [Is RLHF More Difficult than Standard RL?](https://mlsanity.com/paper/2306.14111/) (2023)
- [Feature Engineering for Predictive Modeling using Reinforcement Learning](https://mlsanity.com/paper/1709.07150/) (2017)
- [The autofeat Python Library for Automated Feature Engineering and Selection](https://mlsanity.com/paper/1901.07329/) (2019)
- [MixLLM: Dynamic Routing in Mixed Large Language Models](https://mlsanity.com/paper/2502.18482/) (2025)
- [OpenFE: Automated Feature Generation with Expert-level Performance](https://mlsanity.com/paper/2211.12507/) (2022)
- [Augmenting Interpretable Models with LLMs during Training](https://mlsanity.com/paper/2209.11799/) (2022)
- [Group-wise Reinforcement Feature Generation for Optimal and Explainable Representation Space Reconstruction](https://mlsanity.com/paper/2205.14526/) (2022)
- [Self-optimizing Feature Generation via Categorical Hashing Representation and Hierarchical Reinforcement Crossing](https://mlsanity.com/paper/2309.04612/) (2023)
- [Evolutionary Large Language Model for Automated Feature Transformation](https://mlsanity.com/paper/2405.16203/) (2024)
- [Dynamic and Adaptive Feature Generation with LLM](https://mlsanity.com/paper/2406.03505/) (2024)
- [LLM-Enhanced User-Item Interactions: Leveraging Edge Information for Optimized Recommendations](https://mlsanity.com/paper/2402.09617/) (2024)
- [Reinforcement-Enhanced Autoregressive Feature Transformation: Gradient-steered Search in Continuous Space for Postfix Expressions](https://mlsanity.com/paper/2309.13618/) (2023)
- [Unsupervised Generative Feature Transformation via Graph Contrastive Pre-training and Multi-objective Fine-tuning](https://mlsanity.com/paper/2405.16879/) (2024)
- [Knockoff-Guided Feature Selection via A Single Pre-trained Reinforced Agent](https://mlsanity.com/paper/2403.04015/) (2024)
- [Neuro-Symbolic Embedding for Short and Effective Feature Selection via Autoregressive Generation](https://mlsanity.com/paper/2404.17157/) (2024)
- [A Survey on Data-Centric AI: Tabular Learning from Reinforcement Learning and Generative AI Perspective](https://mlsanity.com/paper/2502.08828/) (2025)
- [Reinforcement Feature Transformation for Polymer Property Performance Prediction](https://mlsanity.com/paper/2409.15616/) (2024)
- [Topology-aware Reinforcement Feature Space Reconstruction for Graph Data](https://mlsanity.com/paper/2411.05742/) (2024)

## Built on by (most cited first, 11 of 11)

- [The FM Agent](https://mlsanity.com/paper/2510.26144/) (2025)
- [Efficient Post-Training Refinement of Latent Reasoning in Large Language Models](https://mlsanity.com/paper/2506.08552/) (2025)
- [Teaching Time Series to See and Speak: Forecasting with Aligned Visual and Textual Perspectives](https://mlsanity.com/paper/2506.24124/) (2025)
- [Autonomous Data Agents: A New Opportunity for Smart Data](https://mlsanity.com/paper/2509.18710/) (2025)
- [LLM-ML Teaming: Integrated Symbolic Decoding and Gradient Search for Valid and Stable Generative Feature Transformation](https://mlsanity.com/paper/2506.09085/) (2025)
- [Bridging the Domain Gap in Equation Distillation with Reinforcement Feedback](https://mlsanity.com/paper/2505.15572/) (2025)
- [Distribution Shift Aware Neural Tabular Learning](https://mlsanity.com/paper/2508.19486/) (2025)
- [StorageXTuner: An LLM Agent-Driven Automatic Tuning Framework for Heterogeneous Storage Systems](https://mlsanity.com/paper/2510.25017/) (2025)
- [Dataforge: Agentic Platform for Autonomous Data Engineering](https://mlsanity.com/paper/2511.06185/) (2025)
- [Evolving Demonstration Optimization for Chain-of-Thought Feature Transformation](https://mlsanity.com/paper/2603.09987/) (2026)
- [LATTEArena: An Evaluation Framework for LLM-powered Tabular Feature Engineering (Extended Version)](https://mlsanity.com/paper/2606.09004/) (2026)

## Abstract (arXiv)

Feature transformation involves generating a new set of features from the original dataset to enhance the data's utility. In certain domains like material performance screening, dimensionality is large and collecting labels is expensive and lengthy. It highly necessitates transforming feature spaces efficiently and without supervision to enhance data readiness and AI utility. However, existing methods fall short in efficient navigation of a vast space of feature combinations, and are mostly designed for supervised settings. To fill this gap, our unique perspective is to leverage a generator-critic duet-play teaming framework using LLM agents and in-context learning to derive pseudo-supervision from unsupervised data. The framework consists of three interconnected steps: (1) Critic agent diagnoses data to generate actionable advice, (2) Generator agent produces tokenized feature transformations guided by the critic's advice, and (3) Iterative refinement ensures continuous improvement through feedback between agents. The generator-critic framework can be generalized to human-agent collaborative generation, by replacing the critic agent with human experts. Extensive experiments demonstrate that the proposed framework outperforms even supervised baselines in feature transformation efficiency, robustness, and practical applicability across diverse datasets.