# MMAI Gym for Science: Training Liquid Foundation Models for Drug Discovery

arXiv: https://arxiv.org/abs/2603.03517 · HTML: https://mlsanity.com/paper/2603.03517/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 35 of its 61 references that are in the map.
- First posted on arXiv 3 Mar 2026.

## Builds on (most cited first, 30 of 35)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Transformers are RNNs: Fast Autoregressive Transformers with Linear Attention](https://mlsanity.com/paper/2006.16236/) (2020)
- [GLU Variants Improve Transformer](https://mlsanity.com/paper/2002.05202/) (2020)
- [Root Mean Square Layer Normalization](https://mlsanity.com/paper/1910.07467/) (2019)
- [GQA: Training Generalized Multi-Query Transformer Models from Multi-Head Checkpoints](https://mlsanity.com/paper/2305.13245/) (2023)
- [Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Duality](https://mlsanity.com/paper/2405.21060/) (2024)
- [Galactica: A Large Language Model for Science](https://mlsanity.com/paper/2211.09085/) (2022)
- [Gated Delta Networks: Improving Mamba2 with Delta Rule](https://mlsanity.com/paper/2412.06464/) (2024)
- [Phi-4-Mini Technical Report: Compact yet Powerful Multimodal Language Models via Mixture-of-LoRAs](https://mlsanity.com/paper/2503.01743/) (2025)
- [Molecular Sets (MOSES): A Benchmarking Platform for Molecular Generation Models](https://mlsanity.com/paper/1811.12823/) (2018)
- [Therapeutics Data Commons: Machine Learning Datasets and Tasks for Drug Discovery and Development](https://mlsanity.com/paper/2102.09548/) (2021)
- [GEOM: Energy-annotated molecular conformations for property prediction and molecular generation](https://mlsanity.com/paper/2006.05531/) (2020)
- [Pocket2Mol: Efficient Molecular Sampling Based on 3D Protein Pockets](https://mlsanity.com/paper/2205.07249/) (2022)
- [Translation between Molecules and Natural Language](https://mlsanity.com/paper/2204.11817/) (2022)
- [LlaSMol: Advancing Large Language Models for Chemistry with a Large-Scale, Comprehensive, High-Quality Instruction Tuning Dataset](https://mlsanity.com/paper/2402.09391/) (2024)
- [BioT5: Enriching Cross-modal Integration in Biology with Chemical Knowledge and Natural Language Associations](https://mlsanity.com/paper/2310.07276/) (2023)
- [Nemotron-H: A Family of Accurate and Efficient Hybrid Mamba-Transformer Models](https://mlsanity.com/paper/2504.03624/) (2025)
- [Retrosynthetic reaction prediction using neural sequence-to-sequence models](https://mlsanity.com/paper/1706.01643/) (2017)
- [MolDiff: Addressing the Atom-Bond Inconsistency Problem in 3D Molecule Diffusion Generation](https://mlsanity.com/paper/2305.07508/) (2023)
- [Training a Scientific Reasoning Model for Chemistry](https://mlsanity.com/paper/2506.17238/) (2025)
- [TxGemma: Efficient and Agentic LLMs for Therapeutics](https://mlsanity.com/paper/2504.06196/) (2025)
- [Nature Language Model: Deciphering the Language of Nature for Scientific Discovery](https://mlsanity.com/paper/2502.07527/) (2025)
- [Pre-RMSNorm and Pre-CRMSNorm Transformers: Equivalent and Efficient Pre-LN Transformers](https://mlsanity.com/paper/2305.14858/) (2023)
- [nach0: Multimodal Natural and Chemical Languages Foundation Model](https://mlsanity.com/paper/2311.12410/) (2023)
- [L+M-24: Building a Dataset for Language + Molecules @ ACL 2024](https://mlsanity.com/paper/2403.00791/) (2024)
- [A Fine-tuning Dataset and Benchmark for Large Language Models for Protein Understanding](https://mlsanity.com/paper/2406.05540/) (2024)
- [GeLLMO: Generalizing Large Language Models for Multi-property Molecule Optimization](https://mlsanity.com/paper/2502.13398/) (2025)
- [BindGPT: A Scalable Framework for 3D Molecular Design via Language Modeling and Reinforcement Learning](https://mlsanity.com/paper/2406.03686/) (2024)

## Abstract (arXiv)

General-purpose large language models (LLMs) that rely on in-context learning do not reliably deliver the scientific understanding and performance required for drug discovery tasks. Simply increasing model size or introducing reasoning tokens does not yield significant performance gains. To address this gap, we introduce the MMAI Gym for Science, a one-stop shop molecular data formats and modalities as well as task-specific reasoning, training, and benchmarking recipes designed to teach foundation models the 'language of molecules' in order to solve practical drug discovery problems. We use MMAI Gym to train an efficient Liquid Foundation Model (LFM) for these applications, demonstrating that smaller, purpose-trained foundation models can outperform substantially larger general-purpose or specialist models on molecular benchmarks. Across essential drug discovery tasks - including molecular optimization, ADMET property prediction, retrosynthesis, drug-target activity prediction, and functional group reasoning - the resulting model achieves near specialist-level performance and, in the majority of settings, surpasses larger models, while remaining more efficient and broadly applicable in the domain.