SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2502.19980 · ICLR · 2025

Can Textual Gradient Work in Federated Learning?

Xiaoxiao Li, Minghui Chen, Han Yu, Zhi Huang, Ruinan Jin, Wenlong Deng, Yuanyuan Chen

arXiv · PDF · Open in the Atlas

Code that ran

6 of 12 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (1 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 2 violated it and 1 has no contract verdict; 3 failures of our instrument; 6 no recorded run), measured 2026-11-22; 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/2502.19980.json · badge: /api/ran/2502.19980.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.

RepositoryRoleRan
ubc-tea/FedTextGrad marked official in Papers with Code
Papers with Code archive
6 ran · 6 no recorded run
FunctionStatusWhere it lives
SingletonBackwardEngine Ran
ran on a synthesized input · constructed an object; no result was computed
ubc-tea/FedTextGrad/textgrad/optimizer/optimizer.py
pointer only (licence: NONE) · get_code("5008a349531d9698")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ubc-tea/FedTextGrad 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.

_check_if_multimodal Ran
ran, violated the contract we drafted
ubc-tea/FedTextGrad/textgrad/optimizer/optimizer.py
pointer only (licence: NONE) · get_code("6c6b64c6f227a73e")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ubc-tea/FedTextGrad 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.

construct_tgd_prompt Ran
ran; our contract draft was wrong, not the code
ubc-tea/FedTextGrad/textgrad/optimizer/optimizer.py
pointer only (licence: NONE) · get_code("792feffeafa114cd")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ubc-tea/FedTextGrad 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.

get_gradient_and_context_text Ran
ran; our contract draft was wrong, not the code
ubc-tea/FedTextGrad/textgrad/optimizer/optimizer.py
pointer only (licence: NONE) · get_code("ce12de8593347921")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ubc-tea/FedTextGrad 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.

is_valid_url Ran
ran, violated the contract we drafted
ubc-tea/FedTextGrad/textgrad/optimizer/optimizer.py
pointer only (licence: NONE) · get_code("a5790751e1d60937")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ubc-tea/FedTextGrad 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.

validate_engine_or_get_default Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
ubc-tea/FedTextGrad/textgrad/optimizer/optimizer.py
pointer only (licence: NONE) · get_code("1bebe4ea01268cce")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ubc-tea/FedTextGrad 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.

EngineLM Not yet run
unverified (harvested, no recorded run)
ubc-tea/FedTextGrad/textgrad/optimizer/optimizer.py
pointer only (licence: NONE) · get_code("4742d5b210b29ae7")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ubc-tea/FedTextGrad 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.

Optimizer Not yet run
unverified (harvested, no recorded run)
ubc-tea/FedTextGrad/textgrad/optimizer/optimizer.py
pointer only (licence: NONE) · get_code("564ea28ae44d788f")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ubc-tea/FedTextGrad 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.

TextualGradientDescent Not yet run
unverified (harvested, no recorded run)
ubc-tea/FedTextGrad/textgrad/optimizer/optimizer.py
pointer only (licence: NONE) · get_code("37c7ec07ab37ca27")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ubc-tea/FedTextGrad 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.

_backward_idempotent Not yet run
unverified (harvested, no recorded run)
ubc-tea/FedTextGrad/textgrad/optimizer/optimizer.py
pointer only (licence: NONE) · get_code("6facfd654eb7db2e")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ubc-tea/FedTextGrad 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.

_check_and_reduce_gradients Not yet run
unverified (harvested, no recorded run)
ubc-tea/FedTextGrad/textgrad/optimizer/optimizer.py
pointer only (licence: NONE) · get_code("b40994628dd26d87")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ubc-tea/FedTextGrad 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.

get_engine Not yet run
unverified (harvested, no recorded run)
ubc-tea/FedTextGrad/textgrad/optimizer/optimizer.py
pointer only (licence: NONE) · get_code("d0b6421efc2f659d")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, ubc-tea/FedTextGrad 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.

Also present in other papers' code

Of 12 samples lifted from this paper's linked repositories, 12 carry a function identity and 0 do not. Of the 12 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 12 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.

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 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Recent studies highlight the promise of LLM-based prompt optimization, especially with TextGrad (Yuksekgonul et al., 2024), which automates "differentiation" via texts and backpropagates textual feedback provided by LLMs. This approach facilitates training in various real-world applications that do not support numerical gradient propagation or loss calculation. It opens new avenues for optimization in decentralized, resource-constrained environments, suggesting that users of black-box LLMs (e.g., ChatGPT) could enhance components of LLM agentic systems (such as prompt optimization) through collaborative paradigms like federated learning (FL). In this paper, we systematically explore the potential and challenges of incorporating textual gradient into FL. Our contributions are fourfold. Firstly, we introduce a novel FL paradigm, Federated Textual Gradient (FedTextGrad), that allows FL clients to upload their locally optimized prompts derived from textual gradients, while the FL server aggregates the received prompts through text summarization. Unlike traditional FL frameworks, which are designed for numerical aggregation, FedTextGrad is specifically tailored for handling textual data, expanding the applicability of FL to a broader range of problems that lack well-defined numerical loss functions. Secondly, building on this design, we conduct extensive experiments to explore the feasibility of federated textual gradients. Our findings highlight the importance of properly tuning key factors (e.g., local steps) in FL training to effectively integrate textual gradients. Thirdly, We highlight a major challenge in federated textual gradient aggregation: retaining essential information from distributed prompt updates. Concatenation often produces prompts that exceed the LLM API's context window, while summarization can degrade performance by generating overly condensed or complex text that lacks key context. Last but not least, in response to this issue, we improve the vanilla variant of FedTextGrad by providing actionable guidance to the LLM when summarizing client prompts by leveraging the Uniform Information Density principle. Such a design reduces the complexity of the aggregated global prompt, thereby better incentivizing the LLM's reasoning ability. Through this principled study, we enable the adoption of textual gradients in FL for optimizing LLMs, identify important issues, and pinpoint future directions, thereby opening up a new research area that warrants further investigation. Our code is available on https://github.com/ubc-tea/FedTextGrad.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2502.19980")
get_code_for_paper("2502.19980")
have("2502.19980")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2502.19980.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2502.19980.svg)](https://syntology.ai/paper/2502.19980)

Connect an agent — have() is free.