# Contrastive MIM: A Contrastive Mutual Information Framework for Unified Generative and Discriminative Representation Learning

arXiv: https://arxiv.org/abs/2502.19642 · HTML: https://mlsanity.com/paper/2502.19642/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 24 of its 36 references that are in the map.
- First posted on arXiv 27 Feb 2025.

## Builds on (most cited first, 24 of 24)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Representation Learning with Contrastive Predictive Coding](https://mlsanity.com/paper/1807.03748/) (2018)
- [Momentum Contrast for Unsupervised Visual Representation Learning](https://mlsanity.com/paper/1911.05722/) (2019)
- [Bootstrap your own latent: A new approach to self-supervised Learning](https://mlsanity.com/paper/2006.07733/) (2020)
- [Exploring Simple Siamese Representation Learning](https://mlsanity.com/paper/2011.10566/) (2020)
- [Understanding Contrastive Representation Learning through Alignment and Uniformity on the Hypersphere](https://mlsanity.com/paper/2005.10242/) (2020)
- [Learning deep representations by mutual information estimation and maximization](https://mlsanity.com/paper/1808.06670/) (2018)
- [Perceiver: General Perception with Iterative Attention](https://mlsanity.com/paper/2103.03206/) (2021)
- [InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets](https://mlsanity.com/paper/1606.03657/) (2016)
- [MiniCPM: Unveiling the Potential of Small Language Models with Scalable Training Strategies](https://mlsanity.com/paper/2404.06395/) (2024)
- [MedMNIST v2 -- A large-scale lightweight benchmark for 2D and 3D biomedical image classification](https://mlsanity.com/paper/2110.14795/) (2021)
- [NV-Embed: Improved Techniques for Training LLMs as Generalist Embedding Models](https://mlsanity.com/paper/2405.17428/) (2024)
- [Keystroke Dynamics as Part of Lifelogging](https://mlsanity.com/paper/2010.10144/) (2020)
- [NeMo: a toolkit for building AI applications using Neural Modules](https://mlsanity.com/paper/1909.09577/) (2019)
- [EMNIST: an extension of MNIST to handwritten letters](https://mlsanity.com/paper/1702.05373/) (2017)
- [InfoVAE: Information Maximizing Variational Autoencoders](https://mlsanity.com/paper/1706.02262/) (2017)
- [Generalized Denoising Auto-Encoders as Generative Models](https://mlsanity.com/paper/1305.6663/) (2013)
- [Improving Small Molecule Generation using Mutual Information Machine](https://mlsanity.com/paper/2208.09016/) (2022)
- [Similar Data Points Identification with LLM: A Human-in-the-loop Strategy Using Summarization and Hidden State Insights](https://mlsanity.com/paper/2404.04281/) (2024)
- [SentenceMIM: A Latent Variable Language Model](https://mlsanity.com/paper/2003.02645/) (2020)
- [MIM: Mutual Information Machine](https://mlsanity.com/paper/1910.03175/) (2019)

## Abstract (arXiv)

Learning representations that generalize well to unknown downstream tasks is a central challenge in representation learning. Existing approaches such as contrastive learning, self-supervised masking, and denoising auto-encoders address this challenge with varying trade-offs. In this paper, we introduce the {contrastive Mutual Information Machine} (cMIM), a probabilistic framework that augments the Mutual Information Machine (MIM) with a novel contrastive objective. While MIM maximizes mutual information between inputs and latent variables and encourages clustering of latent codes, its representations underperform on discriminative tasks compared to state-of-the-art alternatives. cMIM addresses this limitation by enforcing global discriminative structure while retaining MIM's generative strengths. We present two main contributions: (1) we propose cMIM, a contrastive extension of MIM that eliminates the need for positive data augmentation and is robust to batch size, unlike InfoNCE-based methods; (2) we introduce {informative embeddings}, a general technique for extracting enriched representations from encoder--decoder models that substantially improve discriminative performance without additional training, and which apply broadly beyond MIM. Empirical results demonstrate that cMIM consistently outperforms MIM and InfoNCE in classification and regression tasks, while preserving comparable reconstruction quality. These findings suggest that cMIM provides a unified framework for learning representations that are simultaneously effective for discriminative and generative applications.