# Overconfident Oracles: Limitations of In Silico Sequence Design Benchmarking

arXiv: https://arxiv.org/abs/2502.17246 · HTML: https://mlsanity.com/paper/2502.17246/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 31 references that are in the map.
- First posted on arXiv 24 Feb 2025.

## Builds on (most cited first, 13 of 13)

- [Evaluating Protein Transfer Learning with TAPE](https://mlsanity.com/paper/1906.08230/) (2019)
- [Biological Sequence Design with GFlowNets](https://mlsanity.com/paper/2203.04115/) (2022)
- [Conditioning by adaptive sampling for robust design](https://mlsanity.com/paper/1901.10060/) (2019)
- [Design-Bench: Benchmarks for Data-Driven Offline Model-Based Optimization](https://mlsanity.com/paper/2202.08450/) (2022)
- [Conservative Objective Models for Effective Offline Model-Based Optimization](https://mlsanity.com/paper/2107.06882/) (2021)
- [Model Inversion Networks for Model-Based Optimization](https://mlsanity.com/paper/1912.13464/) (2019)
- [Autofocused oracles for model-based design](https://mlsanity.com/paper/2006.08052/) (2020)
- [Protein Discovery with Discrete Walk-Jump Sampling](https://mlsanity.com/paper/2306.12360/) (2023)
- [Population-Based Black-Box Optimization for Biological Sequence Design](https://mlsanity.com/paper/2006.03227/) (2020)
- [Bootstrapped Training of Score-Conditioned Generator for Offline Design of Biological Sequences](https://mlsanity.com/paper/2306.03111/) (2023)
- [Bidirectional Learning for Offline Model-based Biological Sequence Design](https://mlsanity.com/paper/2301.02931/) (2023)
- [Importance Weighted Expectation-Maximization for Protein Sequence Design](https://mlsanity.com/paper/2305.00386/) (2023)
- [Antibody DomainBed: Out-of-Distribution Generalization in Therapeutic Protein Design](https://mlsanity.com/paper/2407.21028/) (2024)

## Built on by (most cited first, 3 of 3)

- [ScienceFlow: A long-horizon agent for ML research, scientific discovery and beyond](https://mlsanity.com/paper/2608.14354/) (2026)
- [ProSpero: Active Learning for Robust Protein Design Beyond Wild-Type Neighborhoods](https://mlsanity.com/paper/2505.22494/) (2025)
- [Search at the Cost of Sampling: Nearly-Instant Latent Space Bayesian Optimization](https://mlsanity.com/paper/2609.19476/) (2026)

## Abstract (arXiv)

Machine learning methods can automate the in silico design of biological sequences, aiming to reduce costs and accelerate medical research. Given the limited access to wet labs, in silico design methods commonly use an oracle model to evaluate de novo generated sequences. However, the use of different oracle models across methods makes it challenging to compare them reliably, motivating the question: are in silico sequence design benchmarks reliable? In this work, we examine 12 sequence design methods that utilise ML oracles common in the literature and find that there are significant challenges with their cross-consistency and reproducibility. Indeed, oracles differing by architecture, or even just training seed, are shown to yield conflicting relative performance with our analysis suggesting poor out-of-distribution generalisation as a key issue. To address these challenges, we propose supplementing the evaluation with a suite of biophysical measures to assess the viability of generated sequences and limit out-of-distribution sequences the oracle is required to score, thereby improving the robustness of the design procedure. Our work aims to highlight potential pitfalls in the current evaluation process and contribute to the development of robust benchmarks, ultimately driving the improvement of in silico design methods.