FlowingFlowing
← Back to all development cases
Development case · blind model judgment pending independent human review; not a formal Benchmark conclusion.
AutocompleteEvidence version preferred20 / 162 · 525faa866ad5b626

Generative AI for Controllable Protein Sequence Design: A Survey

Quantitative biology · 2402.10516v1

FLOWING EVIDENCE BENCHMARK

How do we tell whether Evidence helps?

At the same writing position, with the same model and task, how does supplying retrieved paper passages change the first output? We compare matched versions and retain ties, unusable outputs and incomplete reviews.

SAME DRAFT · EVIDENCE ON OR OFF

01 · FIXED WRITING POSITION

MANUSCRIPT

Same writing position ▌

Matched autocomplete at the same draft position

02 · TWO MATCHED INPUTS

SHARED BY BOTH

Manuscript context, model, task and prompt

A · With Evidence

Retrieved paper passages supplied

B · Without Evidence

No retrieved passages supplied

LLM

Same model and version

A → first output

B → first output

Blind judge agent

First outputs are anonymized as X and Y

Continuation review: accuracy, fit to the writing task and usability
Returns: X preferred / tie / Y preferred / both unusable

Order check: X / Y → Y / X

HOW DOES BLIND JUDGING WORK?

① Anonymize both outputs
The judge sees the same draft and both first outputs without knowing which received Evidence.

② Compare and swap order
The judge applies task-specific criteria in X/Y and then Y/X order.

③ Review disagreements
A third pass resolves disagreements. Incomplete reviews remain in the denominator.

How are autocomplete positions stratified?

Before seeing generation outcomes, we check whether a retrieved passage contains a specific proposition that directly supports the next writing move. Those positions appear in the left opportunity group; the rest are ordinary positions on the right. We select a balanced sample from admitted papers in each field. The 50/50 split is experimental, not a measure of how often either type occurs in writing.

AUTOCOMPLETE · 110

Biology, statistics and astrophysics

55 positions on each side; statistics uses the ten-paper rerun.

AUTOCOMPLETE · 52

Psychology and climate science

26 positions on each side; psychology includes nine papers and climate science four.

How are the table percentages calculated?

Across five fields, 51 of 81 left-column positions preferred the Evidence version. Ties, pairs where both versions were unusable and incomplete reviews remain in the denominator.

51Evidence version preferred
÷
81All positions in this group
=
63%Evidence preference in this group

Source contribution is a separate review: 16 of 22 Evidence wins entered into source review directly used retrieved papers; another 29 wins await review.

These are development-stage model judgments pending independent human review. They are not formal Benchmark conclusions and do not, on their own, establish causality.

Manuscript writing position

Text excerpt · not a PDF page

Research manuscript · excerpt

Generative AI for Controllable Protein Sequence Design: A Survey

1 Introduction

…t of the protein design space, lay the groundwork for sequence-based protein design. Deep generative models offer a promising tool for capturing the distribution underlying natural protein sequences to generate novel and diverse proteins. Moreover, making protein design controllable is crucial for the transition from theory to practice. Ideally, a practical method ought to generate proteins that adhere to task-specific structural and functional constraints. To this end, researchers have utilized conditional generative models and optimization algorithms to realize such tailored protein design.

…cts of the literature in terms of methodologies and tasks. For instance, compared to de novo design, the equally important theme of mutation-based optimization tends to be overlooked. In contrast, our survey endeavors to bridge this gap. We seek to demystify protein sequence design for machine learning researchers by systematically categorizing various controllable design tasks according to the constraints involved. We hope this survey reaches both the machine learning and biochemistry fields and inspires further collaboration towards developing more sophisticated and controllable algorithms.

↓ Next paragraph was generated here

The highlighted text is near the writing position. This is a layout of the evaluation input, not a PDF screenshot.

A matched comparison at the same writing positionJudgment: Evidence version preferred

Both arms use the same model, task and manuscript context. Only the retrieved passages differ. The first recorded output from each arm is shown below.

Evidence

Same model and position · retrieved passages supplied · first output

Preferred by blind review

Controllability in protein sequence design is typically pursued through two complementary strategies: conditional generative models that steer sampling toward a specified property, and optimization procedures that search the latent or sequence space for high-fitness candidates. Both strategies rely on a learned representation of the sequence distribution, which determines how readily the design space can be explored under task-specific constraints.

Passages supplied to the Evidence version

Importance Weighted Expectation-Maximization for Protein Sequence Design

In this paper, we propose IsEM-Pro, an approach to generate protein sequences towards a given fitness criterion. At its core, IsEM-Pro is a latent generative model, augmented by combinatorial structure features from a separately learned Markov random fields (MRFs). We develop an Monte Carlo Expectat…

Read full passage excerpt

In this paper, we propose IsEM-Pro, an approach to generate protein sequences towards a given fitness criterion. At its core, IsEM-Pro is a latent generative model, augmented by combinatorial structure features from a separately learned Markov random fields (MRFs). We develop an Monte Carlo Expectation-Maximization method (MCEM) to learn the model. During inference, sampling from its latent space enhances diversity while its MRFs features guide the exploration in high fitness regions. Experiments on eight protein sequence design tasks show that our IsEM-Pro outperforms the previous best methods by at least 55% on average fitness score and generates more diverse and novel protein sequences.

No Evidence

Same model and position · no retrieved passages · first output

Not preferred in this comparison

No continuation text was produced.

Model action：complete

Blind model judgment

The blinded model review preferred the Evidence version. Its original rationale appears below.

Read the original model judgment

Output A provides a coherent academic paragraph that continues the manuscript's discussion of controllable protein design, bridging from the general introduction to the survey's scope. It is grounded in the draft's themes of conditional generative models and optimization algorithms for task-specific constraints, without introducing unsupported claims or citation markers. Output B is empty despite the task clearly requiring content for the blank paragraph, making it unusable.

The passages are input material; their presence does not mean an output used them or that they support every claim. Draft and source passages are excerpted; no simulated scores or PDF appear here.