Chenguang Zhu, Jerry Li, Michael Zeng, Yin Lee, Dan Iter, Reid Pryzant
5 of 9 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (2 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 2 have no contract verdict; 3 failures of our instrument; 4 no recorded run), measured 2026-11-21; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.
"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2305.03495.json · badge: /api/ran/2305.03495.svg
Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.
| Repository | Role | Ran |
|---|---|---|
| microsoft/lmops | marked official in Papers with Code Papers with Code archive |
2 ran · 2 no recorded run |
| seasonyao/automatic_prompt_optimization_physician_prompting | community (classifier label) classifier basis not recorded on the link |
2 ran · 1 no recorded run |
| zcrwind/prefer | community Papers with Code archive |
1 ran · 1 no recorded run |
| Function | Status | Where it lives |
|---|---|---|
| BaseModel | Ran ran on a synthesized input · constructed an object; no result was computed |
zcrwind/prefer/src/ptuning.py code served (permissive licence) · get_code("e980abcb95fe131a") How it ranRan in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, BaseModel: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.
|
| ProTeGi | Ran ran on a synthesized input · constructed an object; no result was computed |
microsoft/lmops/prompt_optimization/optimizers.py code served (permissive licence) · get_code("9568c0c898993622") How it ranRan in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned an instance of its own class, ProTeGi: the run constructed the object and did not call it, so no result was computed. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.
|
| make_triples | Ran ran; our contract draft was wrong, not the code |
seasonyao/automatic_prompt_optimization_physician_prompting/src/APO_Medical_Prompting/metrics.py pointer only (licence: Apache-2.0); the stored copy get_code serves, from seasonyao/LearnFromHumanEdit, is pointer only (licence: NONE) · get_code("ae0cdcbf80c23151") How it ranThe body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from seasonyao/LearnFromHumanEdit, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded. |
| parse_sectioned_prompt | Ran ran; our contract draft was wrong, not the code |
microsoft/lmops/prompt_optimization/optimizers.py code served (permissive licence) · get_code("37e7baf06cc5b2f0") How it ranRan in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned dict. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.
|
| process_triples | Ran ran; our contract draft was wrong, not the code |
seasonyao/automatic_prompt_optimization_physician_prompting/src/APO_Medical_Prompting/metrics.py pointer only (licence: Apache-2.0); the stored copy get_code serves, from seasonyao/LearnFromHumanEdit, is pointer only (licence: NONE) · get_code("814172c8e37b1334") How it ranThe body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from seasonyao/LearnFromHumanEdit, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded. |
| PromptOptimizer | Not yet run unverified (harvested, no recorded run) |
microsoft/lmops/prompt_optimization/optimizers.py code served (permissive licence) · get_code("f3b032f2bc5e6cc4") How it ranno invocation recorded. |
| RoBERTaVTuningClassification | Not yet run unverified (harvested, no recorded run) |
zcrwind/prefer/src/ptuning.py code served (permissive licence) · get_code("9295971639a2393d") How it ranno invocation recorded. |
| chatgpt | Not yet run unverified (harvested, no recorded run) |
microsoft/lmops/prompt_optimization/optimizers.py code served (permissive licence) · get_code("4fa6d30b6afe3ef8") How it ranno invocation recorded. |
| remove_stopword_and_punc_in_list | Not yet run unverified (harvested, no recorded run) |
seasonyao/automatic_prompt_optimization_physician_prompting/src/APO_Medical_Prompting/metrics.py pointer only (licence: Apache-2.0); the stored copy get_code serves, from seasonyao/LearnFromHumanEdit, is pointer only (licence: NONE) · get_code("a82a905574e3548e") How it ranThe body of this sample is withheld: this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from seasonyao/LearnFromHumanEdit, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded. |
Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 distinct functions with an identity, 6 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 4 later) and 3 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).
Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.
| Function | Also present in | Earliest-dated | Others named |
|---|---|---|---|
| BaseModel brick of this paper's assembly: model |
2 other papers 2 dated earlier · 1 only as an assembly brick |
2012.15723 (2020-12) Making Pre-trained Language Models Better Few-shot Learners |
2212.09257 (2022-12) PromptBoosting: Black-Box Text Classification with Ten Forward Passes |
| remove_stopword_and_punc_in_list | 2 other papers 2 later |
2310.05857 (2023-10) Improving Summarization with Human Edits |
2310.20033 (2023-10) Synthetic Imitation Edit Feedback for Factual Alignment in Clinical Summarization |
| RoBERTaVTuningClassification brick of this paper's assembly: entry |
1 other paper 1 dated earlier |
2212.09257 (2022-12) PromptBoosting: Black-Box Text Classification with Ten Forward Passes |
|
| make_triples | 1 other paper 1 later |
2310.20033 (2023-10) Synthetic Imitation Edit Feedback for Factual Alignment in Clinical Summarization |
|
| parse_sectioned_prompt brick of this paper's assembly: helper |
1 other paper 1 later |
2410.14826 (2024-10) SPRIG: Improving Large Language Model Performance by System Prompt Optimization |
|
| process_triples | 1 other paper 1 later |
2310.20033 (2023-10) Synthetic Imitation Edit Feedback for Factual Alignment in Clinical Summarization |
Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-21.
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
Large Language Models (LLMs) have shown impressive performance as general purpose agents, but their abilities remain highly dependent on prompts which are hand written with onerous trial-and-error effort. We propose a simple and nonparametric solution to this problem, Prompt Optimization with Textual Gradients (ProTeGi), which is inspired by numerical gradient descent to automatically improve prompts, assuming access to training data and an LLM API. The algorithm uses minibatches of data to form natural language "gradients" that criticize the current prompt, much like how numerical gradients point in the direction of error ascent. The natural language gradients are then "propagated" into the prompt by editing the prompt in the opposite semantic direction of the gradient. These gradient descent steps are guided by a beam search and bandit selection procedure which significantly improves algorithmic efficiency. Preliminary results across three benchmark NLP tasks and the novel problem of LLM jailbreak detection suggest that Automatic Prompt Optimization can outperform prior prompt editing techniques and improve an initial prompt's performance by up to 31%, by using data to rewrite vague task descriptions into more precise annotation instructions. 1
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2305.03495")
get_code_for_paper("2305.03495")
have("2305.03495")
The run record, dated, one paper per request, free:
curl https://syntology.ai/api/ran/2305.03495.json
A badge for a README (the split and the date, never a ratio):
[](https://syntology.ai/paper/2305.03495)
Connect an agent — have() is free.