---
title: "OPERA: Operator-Residual Feedback for Reliable Autonomous Optical Experiments with Language-Model Agents"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2608.05990v1"
permalink: "https://feed.astralor.com/items/itm_5543efc1e43aa7dd6868a724"
published: "2026-08-06T13:03:22.000Z"
score: 72
channel: "tech"
tags: ["OPERA","Optical experiments","Digital twin","Residual feedback","Language-model Agents"]
---

# OPERA: Operator-Residual Feedback for Reliable Autonomous Optical Experiments with Language-Model Agents

## AI commentary

> What stands out most is how this paper dethrones raw scores and uses physical residuals as the agent's compass. For anyone working on automated experiments and AI4Science, it's a key step toward turning language models from chat experts into reliable lab assistants. Especially recommended for teams in optics, quantum measurement, and high-repetition experimental workflows.

## At a glance

Autonomous agents often select actions based on scores that may not reflect experimental success. The OPERA framework represents experimental actions as optical operators and evaluates outcomes with physically interpretable residuals. Across three optical tasks, score-only feedback produced score increases without physical improvement in 23.6%–39.0% of decisions, versus 0.9%–1.9% for operator-residual feedback. It also increased the probability of reaching and maintaining task targets and reduced experimental budgets. Protocols selected in digital twins were transferred to three optical instruments, and a lower projection budget was observed in structured-light reconstruction.

## Key points

- OPERA encodes experimental actions as optical operators and uses physical residuals to measure deviation from expected conditions.
- Compared to score-only feedback, operator-residual feedback reduces ineffective decisions from 23.6%–39.0% to 0.9%–1.9%.
- Protocols derived in digital twins were successfully transferred to three real optical instruments.
- The method significantly reduced projection budgets in structured-light reconstruction.

## Full text

Autonomous agents choose actions using scores that may not reflect experimental success. We developed OPERA, an operator-residual framework for optical experiments. It represents experimental actions as optical operators and evaluates their outcomes using physically interpretable residuals. Operators specify executable changes to measurement, control or reconstruction, while residuals report departures from specified physical conditions. The agent uses both to select, combine or generate operators, and physical performance is evaluated independently against a withheld reference. Across three optical tasks, score-only feedback produced score increases without physical improvement in 23.6--39.0\% of decisions, compared with 0.9--1.9\% for operator-residual feedback. Operator-residual feedback increased the probability of reaching and maintaining task targets and reduced experimental budgets. Protocols selected in digital twins were transferred to three optical instruments, and repeated experiments showed a lower projection budget in structured-light reconstruction. Together, operators and residuals guide autonomous decisions using measurable physical evidence.

---

Verify against the original before citing; curated by Astral Feed: https://feed.astralor.com/items/itm_5543efc1e43aa7dd6868a724