SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2410.20749 · NeurIPS · 2025

Matryoshka Pilot: Learning to Drive Black-Box LLMs with LLMs

Hanjun Dai, Bo Dai, Chao Zhang, Yuchen Zhuang, Haotian Sun, Rushi Qiang, Changhao Li

arXiv · PDF · Open in the Atlas

Code that ran

5 of 6 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 1 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 4 failures of our instrument; 1 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/2410.20749.json · badge: /api/ran/2410.20749.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
lichangh20/Matryoshka canonical (classifier label)
classifier basis: census
5 ran · 1 no recorded run
FunctionStatusWhere it lives
compute_approx_kl Ran
ran, honoured the contract we drafted
lichangh20/Matryoshka/models/utils.py
pointer only (licence: NONE) · get_code("60d73ddf6afc1bc3")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, lichangh20/Matryoshka 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.

log_probs_from_logits Ran
ran; our contract draft was wrong, not the code
lichangh20/Matryoshka/models/actor.py
pointer only (licence: NONE) · get_code("959df7824bf09cc1")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, lichangh20/Matryoshka 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.

masked_mean Ran
ran; our contract draft was wrong, not the code
lichangh20/Matryoshka/models/utils.py
pointer only (licence: NONE) · get_code("a68c73a9ac646069")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, lichangh20/Matryoshka 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_solution Ran
ran; our contract draft was wrong, not the code
lichangh20/Matryoshka/gsm_generate_data.py
pointer only (licence: NONE) · get_code("e8baf512d6e432c8")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, lichangh20/Matryoshka 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.

process_ob Ran
ran; our contract draft was wrong, not the code
lichangh20/Matryoshka/alfworld_generate_data.py
pointer only (licence: NONE) · get_code("dfb17eb3e6eaed35")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, lichangh20/Matryoshka 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.

compute_reward Not yet run
unverified (harvested, no recorded run)
lichangh20/Matryoshka/models/utils.py
pointer only (licence: NONE) · get_code("99a2d91ea5aed978")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, lichangh20/Matryoshka 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 6 samples lifted from this paper's linked repositories, 6 carry a function identity and 0 do not. Of the 6 distinct functions with an identity, 4 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 2 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.

FunctionAlso present inEarliest-datedOthers named
log_probs_from_logits 8 other papers
2 dated earlier, 6 later
2306.00398 (2023-06)
Preference-grounded Token-level Guidance for Language Model Fine-tuning
2404.18922 (2024-04)
DPO Meets PPO: Reinforced Token Optimization for RLHF
2412.01981 (2024-12)
Free Process Rewards without Process Labels
and 5 more not named here
process_ob 6 other papers
4 dated earlier, 2 later
2303.11366 (2023-03)
Reflexion: Language Agents with Verbal Reinforcement Learning
2311.16714 (2023-11)
Embodied Multi-Modal Agent trained by an LLM from a Parallel TextWorld
2402.16181 (2024-02)
How Can LLM Guide RL? A Value-Based Approach
and 3 more not named here
compute_approx_kl 3 other papers
3 later
2412.16145 (2024-12)
Offline Reinforcement Learning for LLM Multi-Step Reasoning
2503.10990 (2025-03)
Statistical Impossibility and Possibility of Aligning LLMs with Human Preferences: From Condorcet Paradox to Nash Equilibrium
2606.09124 (2026-06)
A Regret Minimization Framework on Preference Learning in Large Language Models
compute_reward 3 other papers
3 later
2412.16145 (2024-12)
Offline Reinforcement Learning for LLM Multi-Step Reasoning
2503.10990 (2025-03)
Statistical Impossibility and Possibility of Aligning LLMs with Human Preferences: From Condorcet Paradox to Nash Equilibrium
2606.09124 (2026-06)
A Regret Minimization Framework on Preference Learning in Large Language Models

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

Despite the impressive generative abilities of black-box large language models (LLMs), their inherent opacity hinders further advancements in capabilities such as reasoning, planning, and personalization. Existing works aim to enhance LLM capabilities via domain-specific adaptation, which require additional training on accessible model parameters, an infeasible option for black-box LLMs. To address this challenge, we introduce Matryoshka Pilot (M-Pilot), a lightweight white-box LLM controller that guides a large-scale black-box LLM generator by decomposing complex tasks into a series of intermediate outputs. Specifically, we consider the black-box LLM as an environment, with M-Pilot serving as a policy to provide intermediate guidance through prompts for driving the black-box LLM. M-Pilot is trained to pivot the outputs of the black-box LLM aligning with preferences during iterative interaction, which enables controllable multi-turn generation and self-improvement in optimizing intermediate guidance. Empirical evaluations on diverse tasks demonstrate that our method effectively enhances the capabilities of black-box LLMs in complex, long-horizon tasks. Our code is publicly available at: https://github.com/lichangh20/Matryoshka.

For agents

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

get_harvested_code_for_paper("2410.20749")
get_code_for_paper("2410.20749")
have("2410.20749")

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

curl https://syntology.ai/api/ran/2410.20749.json

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

[![Syntology run record](https://syntology.ai/api/ran/2410.20749.svg)](https://syntology.ai/paper/2410.20749)

Connect an agent — have() is free.