# Unlocking Multimodal Protein Language Models at Inference Time

arXiv: https://arxiv.org/abs/2608.25855 · HTML: https://mlsanity.com/paper/2608.25855/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 19 of its 36 references that are in the map.
- First posted on arXiv 26 Aug 2026.

## Builds on (most cited first, 19 of 19)

- [Neural Discrete Representation Learning](https://mlsanity.com/paper/1711.00937/) (2017)
- [Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters](https://mlsanity.com/paper/2408.03314/) (2024)
- [Generative Flows on Discrete State-Spaces: Enabling Multimodal Flows with Applications to Protein Co-Design](https://mlsanity.com/paper/2402.04997/) (2024)
- [Guiding a Diffusion Model with a Bad Version of Itself](https://mlsanity.com/paper/2406.02507/) (2024)
- [Scaling up Masked Diffusion Models on Text](https://mlsanity.com/paper/2410.18514/) (2024)
- [Simple Guidance Mechanisms for Discrete Diffusion Models](https://mlsanity.com/paper/2412.10193/) (2024)
- [Derivative-Free Guidance in Continuous and Discrete Diffusion Models with Soft Value-Based Decoding](https://mlsanity.com/paper/2408.08252/) (2024)
- [Improved motif-scaffolding with SE(3) flow matching](https://mlsanity.com/paper/2401.04082/) (2024)
- [La-Proteina: Atomistic Protein Generation via Partially Latent Flow Matching](https://mlsanity.com/paper/2507.09466/) (2025)
- [Unifying Tree Search Algorithm and Reward Design for LLM Reasoning: A Survey](https://mlsanity.com/paper/2510.09988/) (2025)
- [RFG: Test-Time Scaling for Diffusion Large Language Model Reasoning with Reward-Free Guidance](https://mlsanity.com/paper/2509.25604/) (2025)
- [Computational Protein Science in the Era of Large Language Models (LLMs)](https://mlsanity.com/paper/2501.10282/) (2025)
- [ProtInvTree: Deliberate Protein Inverse Folding with Reward-guided Tree Search](https://mlsanity.com/paper/2506.00925/) (2025)
- [GLProtein: Global-and-Local Structure Aware Protein Representation Learning](https://mlsanity.com/paper/2506.06294/) (2025)
- [Controlling Repetition in Protein Language Models](https://mlsanity.com/paper/2602.00782/) (2026)
- [HD-Prot: A Protein Language Model for Joint Sequence-Structure Modeling with Continuous Structure Tokens](https://mlsanity.com/paper/2512.15133/) (2025)
- [Enhancing Molecular Property Predictions by Learning from Bond Modelling and Interactions](https://mlsanity.com/paper/2603.00568/) (2026)
- [Geometric Flow Matching for Molecular Conformation Generation via Manifold Decomposition](https://mlsanity.com/paper/2605.25577/) (2026)
- [MotifBench: A standardized protein design benchmark for motif-scaffolding problems](https://mlsanity.com/paper/2502.12479/) (2025)

## Abstract (arXiv)

Multimodal protein language models (pLMs) learn joint protein sequence-structure distributions, and their generation performance should also depend critically on inference-time sampling strategies. Yet prior work has focused more on model training than on how inference-time strategies behave. In this paper, we establish a three-stage investigation framework to empirically study the inference design space of multimodal pLMs across three representative pLMs and four fundamental tasks. We evaluate vanilla sampling, task-specific classifier-free guidance, and reward-guided beam search on multimodal pLMs, corresponding to controls over sampling distributions, per-step logits, and parallel trajectories. Throughout the complementary advancements centered on exploration-exploitation trade-off, we (1) reveal the suboptimality of default inference protocols and identify task-oriented sampling preferences; (2) observe substantial quantitative gains across tasks, consistently boosting the upper bound performance of multimodal pLMs without updating model parameters; (3) derive conclusions about base models that differ from prior consensus.