SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2602.10458 · 2026

Found-RL: foundation model-enhanced reinforcement learning for autonomous driving

Zilin Huang, Tianyi Wang, Yansong Qu, Yuhao Luo, Yiheng Feng, Sikai Chen, Zihao Sheng, Jiancong Chen, Samuel Labi

arXiv · PDF · Open in the Atlas

Code that ran

7 of 8 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; 0 honoured the contract we drafted, 1 violated it and 0 have no contract verdict; 6 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/2602.10458.json · badge: /api/ran/2602.10458.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
ys-qu/found-rl canonical (classifier label)
classifier basis: census
7 ran · 1 no recorded run
FunctionStatusWhere it lives
calculate_metrics Ran
ran; our contract draft was wrong, not the code
ys-qu/found-rl/clean_wandb_data.py
pointer only (licence: NOASSERTION) · get_code("cad6e72086af2eb4")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, ys-qu/found-rl (NOASSERTION). 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.

find_col Ran
ran; our contract draft was wrong, not the code
ys-qu/found-rl/clean_wandb_data.py
pointer only (licence: NOASSERTION) · get_code("68903acdf43fb090")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, ys-qu/found-rl (NOASSERTION). 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.

find_project_root Ran
ran; our contract draft was wrong, not the code
ys-qu/found-rl/utils/os_utils.py
pointer only (licence: NOASSERTION) · get_code("f843ba77dadbf6ab")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, ys-qu/found-rl (NOASSERTION). 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.

getitem Ran
ran; our contract draft was wrong, not the code
ys-qu/found-rl/bc_stage/preprocess.py
pointer only (licence: NOASSERTION) · get_code("3c0b98425e701d0d")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, ys-qu/found-rl (NOASSERTION). 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.

load_qwen_vllm Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
ys-qu/found-rl/utils/vlm_utils.py
pointer only (licence: NOASSERTION) · get_code("782ab451aa23b24c")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, ys-qu/found-rl (NOASSERTION). 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_act Ran
ran, violated the contract we drafted
ys-qu/found-rl/bc_stage/preprocess.py
pointer only (licence: NOASSERTION) · get_code("91b2b62f4f1a92ba")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, ys-qu/found-rl (NOASSERTION). 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.

tolerant_parse_vlm_output Ran
ran; our contract draft was wrong, not the code
ys-qu/found-rl/train_rl_asycn_vlm.py
pointer only (licence: NOASSERTION) · get_code("cc74f997e9bcfc23")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, ys-qu/found-rl (NOASSERTION). 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.

load_qwen Not yet run
unverified (harvested, no recorded run)
ys-qu/found-rl/utils/vlm_utils.py
pointer only (licence: NOASSERTION) · get_code("bfd8050d9914e772")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, ys-qu/found-rl (NOASSERTION). 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 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 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 8 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

Reinforcement Learning (RL) has emerged as a dominant paradigm for end-to-end autonomous driving (AD) with real-time inference. However, RL typically suffers from sample inefficiency and a lack of semantic interpretability in complex scenarios. To mitigate these limitations, Foundation Models (particularly, Vision-Language Models (VLMs)) can be integrated because they offer rich, context-aware knowledge. Yet still, deploying such computationally intensive models within high-frequency multi-environment RL training loops is severely hindered by prohibitive inference latency and the absence of unified integration platforms. To bridge this gap, we present Found-RL, a specialized platform tailored to leverage foundation models to efficiently enhance RL for AD. A core innovation of the proposed platform is its asynchronous batch inference framework, which decouples heavy VLM reasoning from the simulation loop. This design effectively resolves latency bottlenecks, supporting real-time or near-real-time RL learning from VLM feedback. Using the proposed platform, we introduce diverse supervision mechanisms to address domain-specific challenges: we first implement Value-Margin Regularization (VMR) and Advantage-Weighted Action Guidance (AWAG) to effectively distill expert-like VLM action suggestions into the RL policy. Furthermore, for dense supervision, we adopt high-throughput CLIP for reward shaping. We mitigate CLIP's dynamic blindness and probability dilution via Conditional Contrastive Action Alignment, which conditions prompts on discretized speed/command and yields a normalized, margin-based bonus from contextspecific action-anchor scoring. Found-RL delivers an end-to-end pipeline for fine-tuned VLM integration with modular support, and shows that a lightweight RL model with millions of parameters can achieve near-VLM performance compared with billion-parameter VLMs while sustaining real-time inference (~500 FPS). Code, data, and models will be publicly available at https://github.com/ys-qu/found-rl.

For agents

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

get_harvested_code_for_paper("2602.10458")
get_code_for_paper("2602.10458")
have("2602.10458")

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

curl https://syntology.ai/api/ran/2602.10458.json

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

[![Syntology run record](https://syntology.ai/api/ran/2602.10458.svg)](https://syntology.ai/paper/2602.10458)

Connect an agent — have() is free.