Self-correcting Reward Shaping via Language Models for Reinforcement Learning Agents in Games

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Afonso, António, Leite, Iolanda, Sestini, Alessandro, Fuchs, Florian, Tollmar, Konrad, Gisslén, Linus
Format:	Preprint
Publié:	2025
Sujets:	Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

TROFI: Trajectory-Ranked Offline Inverse Reinforcement Learning
par: Sestini, Alessandro, et autres
Publié: (2025)

Using Deep Convolutional Neural Networks to Detect Rendered Glitches in Video Games
par: Ling, Carlos Garcia, et autres
Publié: (2024)

Improving Generalization in Game Agents with Data Augmentation in Imitation Learning
par: Yadgaroff, Derek, et autres
Publié: (2023)

Human-Like Goalkeeping in a Realistic Football Simulation: a Sample-Efficient Reinforcement Learning Approach
par: Sestini, Alessandro, et autres
Publié: (2025)

Real-Time Diffusion Policies for Games: Enhancing Consistency Policies with Q-Ensembles
par: Zhang, Ruoqi, et autres
Publié: (2025)

Reinforcement Learning for High-Level Strategic Control in Tower Defense Games
par: Bergdahl, Joakim, et autres
Publié: (2024)

Leveraging Large Language Models for Efficient Failure Analysis in Game Development
par: Marini, Leonardo, et autres
Publié: (2024)

Towards Better Sample Efficiency in Multi-Agent Reinforcement Learning via Exploration
par: Baghi, Amir, et autres
Publié: (2025)

A Benchmark Environment for Offline Reinforcement Learning in Racing Games
par: Macaluso, Girolamo, et autres
Publié: (2024)

ARMS: Automatic Reward Shaping for Sparse-Reward Multi-Agent Reinforcement Learning
par: Abboud, Elie, et autres
Publié: (2026)

Text2Reward: Reward Shaping with Language Models for Reinforcement Learning
par: Xie, Tianbao, et autres
Publié: (2023)

Highly Efficient Self-Adaptive Reward Shaping for Reinforcement Learning
par: Ma, Haozhe, et autres
Publié: (2024)

Extracting Heuristics from Large Language Models for Reward Shaping in Reinforcement Learning
par: Bhambri, Siddhant, et autres
Publié: (2024)

Think in Games: Learning to Reason in Games via Reinforcement Learning with Large Language Models
par: Liao, Yi, et autres
Publié: (2025)

Principal-Agent Reward Shaping in MDPs
par: Ben-Porat, Omer, et autres
Publié: (2023)

Self-Consistency of the Internal Reward Models Improves Self-Rewarding Language Models
par: Zhou, Xin, et autres
Publié: (2025)

SPEQ: Offline Stabilization Phases for Efficient Q-Learning in High Update-To-Data Ratio Reinforcement Learning
par: Romeo, Carlo, et autres
Publié: (2025)

Reward Generation via Large Vision-Language Model in Offline Reinforcement Learning
par: Lee, Younghwan, et autres
Publié: (2025)

Subgoal-based Reward Shaping to Improve Efficiency in Reinforcement Learning
par: Okudo, Takato, et autres
Publié: (2021)

RLSR: Reinforcement Learning from Self Reward
par: Simonds, Toby, et autres
Publié: (2025)

Learning Reward for Robot Skills Using Large Language Models via Self-Alignment
par: Zeng, Yuwei, et autres
Publié: (2024)

Cooperative Reward Shaping for Multi-Agent Pathfinding
par: Song, Zhenyu, et autres
Publié: (2024)

Self-Rewarding Language Models
par: Yuan, Weizhe, et autres
Publié: (2024)

Model-Based Reinforcement Learning in Discrete-Action Non-Markovian Reward Decision Processes
par: Trapasso, Alessandro, et autres
Publié: (2025)

On the Sample Efficiency of Abstractions and Potential-Based Reward Shaping in Reinforcement Learning
par: Canonaco, Giuseppe, et autres
Publié: (2024)

Shaping Sparse Rewards in Reinforcement Learning: A Semi-supervised Approach
par: Li, Wenyun, et autres
Publié: (2025)

Vision-Language Models are Zero-Shot Reward Models for Reinforcement Learning
par: Rocamonde, Juan, et autres
Publié: (2023)

Multi-Agent Collaborative Reward Design for Enhancing Reasoning in Reinforcement Learning
par: Yang, Pei, et autres
Publié: (2025)

Language Agents with Reinforcement Learning for Strategic Play in the Werewolf Game
par: Xu, Zelai, et autres
Publié: (2023)

Improving Conditional Level Generation using Automated Validation in Match-3 Games
par: Aylagas, Monica Villanueva, et autres
Publié: (2024)

AI for Handball: predicting and explaining the 2024 Olympic Games tournament with Deep Learning and Large Language Models
par: Felice, Florian
Publié: (2024)

TextGames: Learning to Self-Play Text-Based Puzzle Games via Language Model Reasoning
par: Hudi, Frederikus, et autres
Publié: (2025)

PyTAG: Tabletop Games for Multi-Agent Reinforcement Learning
par: Balla, Martin, et autres
Publié: (2024)

Enhancing Inverse Reinforcement Learning through Encoding Dynamic Information in Reward Shaping
par: Zhan, Simon Sinong, et autres
Publié: (2024)

ROLeR: Effective Reward Shaping in Offline Reinforcement Learning for Recommender Systems
par: Zhang, Yi, et autres
Publié: (2024)

Generative Floor Plan Design with LLMs via Reinforcement Learning with Verifiable Rewards
par: Lara, Luis, et autres
Publié: (2026)

Multi-Turn Reinforcement Learning for Tool-Calling Agents with Iterative Reward Calibration
par: Modecrua, Wachiravit, et autres
Publié: (2026)

Reward Shaping for Inference-Time Alignment: A Stackelberg Game Perspective
par: Wang, Haichuan, et autres
Publié: (2026)

Large Language Model Guided Incentive Aware Reward Design for Cooperative Multi-Agent Reinforcement Learning
par: Urgun, Dogan, et autres
Publié: (2026)

MIR: Efficient Exploration in Episodic Multi-Agent Reinforcement Learning via Mutual Intrinsic Reward
par: Chen, Kesheng, et autres
Publié: (2025)

if (filemtime($filename) + $maxlifetime < time()) { unlink($filename); $count++; } } return $count; } /** * A function that is called internally when session data is to be saved. * * @param string $sessId The current session ID * @param string $data The session data to write * * @return bool */ protected function saveSession($sessId, $data): bool { $sessFile = $this->path . '/sess_' . $sessId; if ($handle = fopen($sessFile, 'w')) { $return = false; // Lock the file for exclusive access to avoid issues with multiple // processes writing session simultaneously: if (flock($handle, LOCK_EX)) { $return = fwrite($handle, $data); // Make sure that there's no trailing data by truncating the file to // the correct length: ftruncate($handle, strlen($data)); fflush($handle); flock($handle, LOCK_UN); } fclose($handle); if ($return !== false) { return true; } } // If we got this far, something went wrong with the file output... // It is tempting to throw an exception here, but this code is called // outside of the context of exception handling, so all we can do is // echo a message.

// Anecdotal testing Today and Yesterday seems to indicate destroy() // is called by the garbage collector and everything is good. // Something to keep in mind though. return 0; } /** * Write function that is called when session data is to be saved. * * @param string $sessId The current session ID * @param string $data The session data to write * * @return bool */ public function write($sessId, $data): bool { if ($this->writesDisabled) { return true; } return $this->saveSession($sessId, $data); } /** * A function that is called internally when session data is to be saved. * * @param string $sessId The current session ID * @param string $data The session data to write * * @return bool */ abstract protected function saveSession($sessId, $data): bool; }

* * @return void */ public function writeClose() { // The assumption is that we're using PHP's ext/session. // session_write_close() will actually overwrite $_SESSION with an // empty array on completion -- which leads to a mismatch between what // is in the storage object and $_SESSION. To get around this, we // temporarily reset $_SESSION to an array, and then re-link it to // the storage object. // // Additionally, while you _can_ write to $_SESSION following a // session_write_close() operation, no changes made to it will be // flushed to the session handler. As such, we now mark the storage // object isImmutable. $storage = $this->getStorage(); if (! $storage->isImmutable()) { $_SESSION = $storage->toArray(true); session_write_close(); $storage->fromArray($_SESSION); $storage->markImmutable(); } } /** * Attempt to set the session name * * If the session has already been started, or if the name provided fails * validation, an exception will be raised. * * @param string $name * @return SessionManager * @throws Exception\InvalidArgumentException */ public function setName($name) { if ($this->sessionExists()) { throw new Exception\InvalidArgumentException( 'Cannot set session name after a session has already started'

Environment & details:

GET Data empty

POST Data empty

Files empty

Key	Value
VUFIND_SESSION	"32141a54f5b9e02f3725ec4d088c9d56"
language	"fr"
ui	"standard"

Session

Key	Value
__Laminas	array:2 [`"_REQUEST_ACCESS_TIME" => 1790902560.1972 "_VALID" => array:1 [ "Laminas\Session\Validator\Id" => "32141a54f5b9e02f3725ec4d088c9d56" ]`]
SessionState	Laminas\Stdlib\ArrayObject {#938}
FlashMessenger	Laminas\Stdlib\ArrayObject {#1433}
SessionHelper	Laminas\Stdlib\ArrayObject {#1759}

Server/Request Data

Key	Value
REDIRECT_VUFIND_LOCAL_DIR	"/usr/local/vufind/local"
REDIRECT_HTTPS	"on"
REDIRECT_SSL_TLS_SNI	"info.umar.mx"
REDIRECT_STATUS	"200"
VUFIND_LOCAL_DIR	"/usr/local/vufind/local"
HTTPS	"on"
SSL_TLS_SNI	"info.umar.mx"
HTTP_ACCEPT	"/"
HTTP_USER_AGENT	"Mozilla/5.0 AppleWebKit/537.36 (KHTML, like Gecko; compatible; ClaudeBot/1.0; +claudebot@anthropic.com)"
HTTP_COOKIE	"VUFIND_SESSION=32141a54f5b9e02f3725ec4d088c9d56; language=fr; ui=standard"
HTTP_ACCEPT_ENCODING	"gzip, br, zstd, deflate"
HTTP_HOST	"info.umar.mx"
PATH	"/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/snap/bin"
SERVER_SIGNATURE	"<address>Apache/2.4.66 (Ubuntu) Server at info.umar.mx Port 443</address>\n"
SERVER_SOFTWARE	"Apache/2.4.66 (Ubuntu)"
SERVER_NAME	"info.umar.mx"
SERVER_ADDR	"200.23.222.19"
SERVER_PORT	"443"
REMOTE_ADDR	"216.73.217.142"
DOCUMENT_ROOT	"/usr/local/vufind/public"
REQUEST_SCHEME	"https"
CONTEXT_PREFIX	""
CONTEXT_DOCUMENT_ROOT	"/usr/local/vufind/public"
SERVER_ADMIN	"admin@localhost"
SCRIPT_FILENAME	"/usr/local/vufind/public/index.php"
REMOTE_PORT	"65265"
REDIRECT_URL	""" /Record/\n arxiv_https___arxiv_org_abs_2506_23626/Similar """
GATEWAY_INTERFACE	"CGI/1.1"
SERVER_PROTOCOL	"HTTP/1.1"
REQUEST_METHOD	"GET"
QUERY_STRING	""
REQUEST_URI	"/Record/%0Aarxiv_https___arxiv_org_abs_2506_23626/Similar"
SCRIPT_NAME	"/index.php"
PHP_SELF	"/index.php"
REQUEST_TIME_FLOAT	1790902560.1914
REQUEST_TIME	1790902560

Environment Variables empty

Registered Handlers

0. Whoops\Handler\PrettyPageHandler