Reward Design for RL

RL: Reinforcement Learning

Follow this topic Ask about this topic

Reward design studies how to turn task goals, preferences, and constraints into learning signals, especially when success is sparse or objectives compete. Approaches include potential-based shaping and reward machines, preference and rubric models, visual progress rewards, intrinsic exploration bonuses, and automated reward search using language models. These methods are used in settings from robot manipulation and locomotion to language-model reasoning, tool use, and multi-agent coordination. A central tension is that denser or adaptive rewards can improve credit assignment but also mislead policies, destabilize training, or invite reward hacking, so evaluation must test intended outcomes rather than proxy scores alone.

Latest papers

261 papers44 papers in the last four weeks, up 83% on the four weeks before. 0.3% of all new papers.

Apr 22, 2026

Occupancy Reward Shaping: Improving Credit Assignment for Offline Goal-Conditioned Reinforcement Learning

The temporal lag between actions and their long-term consequences makes credit assignment a challenge when learning goal-directed behaviors from data. Generative world models capture the distribution of future states an agent may visit, indicating that they have captured temporal information. How can that temporal information be extracted to perform credit assignment? In this paper, we formalize how the temporal information stored in world models encodes the underlying geometry of the world. Leveraging optimal transport, we extract this geometry from a learned model of the occupancy measure into a reward function that captures goal-reaching information. Our resulting method, Occupancy Reward Shaping, largely mitigates the problem of credit assignment in sparse reward settings. ORS provably does not alter the optimal policy, yet empirically improves performance by 2.2x across 13 diverse long-horizon locomotion and manipulation tasks. Moreover, we demonstrate the effectiveness of ORS in the real world for controlling nuclear fusion on 3 Tokamak control tasks. Code: https://github.com/aravindvenu7/occupancy_reward_shaping; Website: https://aravindvenu7.github.io/website/ors/

Aravind Venugopal, Jiayu Chen, Xudong Wu +3

Carnegie Mellon University · The University of Hong Kong · INFIFORCE Intelligent Technology +1

Reward Shaping Temporal Credit Assignment Offline RL Sparse-Reward RL Policy Learning Reward Design for RL+1

Apr 20, 2026

OGER: A Robust Offline-Guided Exploration Reward for Hybrid Reinforcement Learning

Recent advancements in Reinforcement Learning with Verifiable Rewards (RLVR) have significantly improved Large Language Model (LLM) reasoning, yet models often struggle to explore novel trajectories beyond their initial policy distribution. While offline teacher guidance and entropy-driven strategies have been proposed to address this, they often lack deep integration or are constrained by the model's inherent capacity. In this paper, we propose OGER (Offline-Guided Exploration Reward), a novel framework that unifies offline teacher guidance and online reinforcement learning through a specialized reward modeling lens. OGER employs multi-teacher collaborative training and constructs an auxiliary exploration reward that leverages both offline trajectories and the model's own entropy to incentivize autonomous exploration. Extensive experiments across mathematical and general reasoning benchmarks demonstrate that OGER consistently outperforms competitive baselines, achieving substantial gains in mathematical reasoning while maintaining robust generalization to out-of-domain tasks. We provide a comprehensive analysis of training dynamics and conduct detailed ablation studies to validate the effectiveness of our entropy-aware reward modulation. Our code is available at https://github.com/ecoli-hit/OGER.git.

Xinyu Ma, Mingzhou Xu, Xuebo Liu +4

Institute of Computing and Intelligence, Harbin Institute of Technology, Shenzhen, China · Hithink RoyalFlush Information Network, Hangzhou, China · Computer and Information Science, University of Macau, Macau, China

Reward Shaping RL for Language Model Reasoning RL Exploration Reinforcement Learning with Verifiable Rewards Reward Design for RL Offline-To-Online RL

Apr 20, 2026

Learning to Correct: Calibrated Reinforcement Learning for Multi-Attempt Chain-of-Thought

State-of-the-art reasoning models utilize long chain-of-thought (CoT) to solve increasingly complex problems using more test-time computation. In this work, we explore a long CoT setting where the model makes up to K successive attempts at solving a problem, in which each attempt is allowed to build on earlier ones after the model receives a hard verifier feedback. This motivates RL methods that can harness per-attempt rewards by carefully weighting individual attempts. We study optimizing the Verification@K reward (the model succeeds by the K-th attempt) and show that naively weighing the attempts by their pass/fail results in biased gradients. We introduce Calibrated Attempt-Level (CAL) GRPO by devising a weighing strategy to obtain unbiased gradients while maintaining small variance. Our theory reveals how incorporating per-attempt rewards influence the training and the eventual Verification@K performance. Experiments, baselines, and ablations on synthetic and real data corroborate our theory and the benefits of CAL-GRPO over vanilla GRPO as well as naive weighting.

Muhammed Emrullah Ildiz, Halil Alperen Gozeten, Ege Onur Taga +1

University of Michigan, Ann Arbor · Google Research

CoT Reasoning RL for Language Model Reasoning Group Relative Policy Optimization Credit Assignment in RL Reinforcement Learning with Verifiable Rewards Reward Design for RL

Apr 17, 2026

Fuzzy Logic Theory-based Adaptive Reward Shaping for Robust Reinforcement Learning (FARS)

Reinforcement learning (RL) often struggles in real-world tasks with high-dimensional state spaces and long horizons, where sparse or fixed rewards severely slow down exploration and cause agents to get trapped in local optima. This paper presents a fuzzy logic based reward shaping method that integrates human intuition into RL reward design. By encoding expert knowledge into adaptive and interpreable terms, fuzzy rules promote stable learning and reduce sensitivity to hyperparameters. The proposed method leverages these properties to adapt reward contributions based on the agent state, enabling smoother transitions between fast motion and precise control in challenging navigation tasks. Extensive simulation results on autonomous drone racing benchmarks show stable learning behavior and consistent task performance across scenarios of increasing difficulty. The proposed method achieves faster convergence and reduced performance variability across training seeds in more challenging environments, with success rates improving by up to approximately 5 percent compared to non fuzzy reward formulations.

Hürkan Şahin, Van Huyen Dang, Erdi Sayar +2

Automatic Control Group (RAT), Paderborn University, 33098 Paderborn, Germany

Reward Shaping Autonomous Drone Racing Robotic RL Fuzzy Logic Robust RL Reward Design for RL

Apr 15, 2026

Enhancing Reinforcement Learning for Radiology Report Generation with Evidence-aware Rewards and Self-correcting Preference Learning

Recent reinforcement learning (RL) approaches have advanced radiology report generation (RRG), yet two core limitations persist: (1) report-level rewards offer limited evidence-grounded guidance for clinical faithfulness; and (2) current methods lack an explicit self-improving mechanism to align with clinical preference. We introduce clinically aligned Evidence-aware Self-Correcting Reinforcement Learning (ESC-RL), comprising two key components. First, a Group-wise Evidence-aware Alignment Reward (GEAR) delivers group-wise, evidence-aware feedback. GEAR reinforces consistent grounding for true positives, recovers missed findings for false negatives, and suppresses unsupported content for false positives. Second, a Self-correcting Preference Learning (SPL) strategy automatically constructs a reliable, disease-aware preference dataset from multiple noisy observations and leverages an LLM to synthesize refined reports without human supervision. ESC-RL promotes clinically faithful, disease-aligned reward and supports continual self-improvement during training. Extensive experiments on two public chest X-ray datasets demonstrate consistent gains and state-of-the-art performance.

Qin Zhou, Guoyan Liang, Qianyi Yang +4

Department of Computer Science and Engineering, ECUST, China · Key Laboratory of Smart Manufacturing in Energy Chemical Process, Ministry of Education, P. R. China · Zhejiang University, Hangzhou, China +1

Radiology Report Generation RL for Generative Models Reward Design for RL Preference-Based RL LLM Self-Correction

Apr 15, 2026

Chain of Uncertain Rewards with Large Language Models for Reinforcement Learning

Designing effective reward functions is a cornerstone of reinforcement learning (RL), yet it remains a challenging and labor-intensive process due to the inefficiencies and inconsistencies inherent in traditional methods. Existing methods often rely on extensive manual design and evaluation steps, which are prone to redundancy and overlook local uncertainties at intermediate decision points. To address these challenges, we propose the Chain of Uncertain Rewards (CoUR), a novel framework that integrates large language models (LLMs) to streamline reward function design and evaluation in RL environments. Specifically, our CoUR introduces code uncertainty quantification with a similarity selection mechanism that combines textual and semantic analyses to identify and reuse the most relevant reward function components. By reducing redundant evaluations and leveraging Bayesian optimization on decoupled reward terms, CoUR enables a more efficient and robust search for optimal reward feedback. We comprehensively evaluate CoUR across nine original environments from IsaacGym and all 20 tasks from the Bidexterous Manipulation benchmark. The experimental results demonstrate that CoUR not only achieves better performance but also significantly lowers the cost of reward evaluations.

Shentong Mo

Carnegie Mellon University

Reward Design for RL LLM-Guided RL Robotic RL

Apr 13, 2026

Robust Optimization for Mitigating Reward Hacking with Correlated Proxies

Designing robust reinforcement learning (RL) agents in the presence of imperfect reward signals remains a core challenge. In practice, agents are often trained with proxy rewards that only approximate the true objective, leaving them vulnerable to reward hacking, where high proxy returns arise from unintended or exploitative behaviors. Recent work formalizes this issue using r-correlation between proxy and true rewards, but existing methods like occupancy-regularized policy optimization (ORPO) optimize against a fixed proxy and do not provide strong guarantees against broader classes of correlated proxies. In this work, we formulate reward hacking as a robust policy optimization problem over the space of all r-correlated proxy rewards. We derive a tractable max-min formulation, where the agent maximizes performance under the worst-case proxy consistent with the correlation constraint. We further show that when the reward is a linear function of known features, our approach can be adapted to incorporate this prior knowledge, yielding both improved policies and interpretable worst-case rewards. Experiments across several environments show that our algorithms consistently outperform ORPO in worst-case returns, and offer improved robustness and stability across different levels of proxy-true reward correlation. These results show that our approach provides both robustness and transparency in settings where reward design is inherently uncertain. The code is available at https://github.com/ZixuanLiu4869/reward_hacking.

Zixuan Liu, Xiaolin Sun, Zizhan Zheng

Department of Computer Science Tulane University New Orleans, LA 70118, USA

Reward Hacking Robust RL Reward Design for RL Distributionally Robust RL

Apr 13, 2026

Utilizing and Calibrating Hindsight Process Rewards via Reinforcement with Mutual Information Self-Evaluation

To overcome the sparse reward challenge in reinforcement learning (RL) for agents based on large language models (LLMs), we propose Mutual Information Self-Evaluation (MISE), an RL paradigm that utilizes hindsight generative self-evaluation as dense reward signals while simultaneously calibrating them against the environmental feedbacks. Empirically, MISE enables an agent to learn autonomously from dense internal rewards supplementing sparse extrinsic signals. Theoretically, our work provides the first formal foundation for the paradigm of generative self-rewarding. We prove that utilizing hindsight self-evaluation rewards is equivalent to minimizing an objective that combines mutual information with a KL divergence term between the policy and a proxy reward policy. This theoretical insight then informs and justifies our calibration step, which actively aligns these rewards with the optimal policy. Extensive experiments show that MISE outperforms strong baselines, enabling open-source LLMs about 7B parameters to achieve performance comparable to GPT-4o on validation without expert supervision.

Jiashu Yao, Heyan Huang, Zeming Liu +1

Beijing Institute of Technology · Beihang University

Reinforcement Learning RL for LLM Agents Reward Design for RL Intrinsic Reward Methods Sparse-Reward RL

Apr 13, 2026

Triviality Corrected Endogenous Reward

Reinforcement learning for open-ended text generation is constrained by the lack of verifiable rewards, necessitating reliance on judge models that require either annotated data or powerful closed-source models. Inspired by recent work on unsupervised reinforcement learning for mathematical reasoning using confidence-based endogenous rewards, we investigate whether this principle can be adapted to open-ended writing tasks. We find that directly applying confidence rewards leads to Triviality Bias: the policy collapses toward high-probability outputs, reducing diversity and meaningful content. We propose TCER (Triviality Corrected Endogenous Reward), which addresses this bias by rewarding the relative information gain between a specialist policy and a generalist reference policy, modulated by a probability-dependent correction mechanism. Across multiple writing benchmarks and model architectures, TCER achieves consistent improvements without external supervision. Furthermore, TCER also transfers effectively to mathematical reasoning, validating the generality of our approach across different generation tasks.

Xinda Wang, Zhengxu Hou, Yangshijie Zhang +6

Peking University · Alibaba Group · Lanzhou University

RL for Language Models Reward Design for RL RL Fine-Tuning RL for Language Model Reasoning

Apr 13, 2026

The Past Is Not Past: Memory-Enhanced Dynamic Reward Shaping

Despite the success of reinforcement learning for large language models, a common failure mode is reduced sampling diversity, where the policy repeatedly generates similar erroneous behaviors. Classical entropy regularization encourages randomness under the current policy, but does not explicitly discourage recurrent failure patterns across rollouts. We propose MEDS, a Memory-Enhanced Dynamic reward Shaping framework that incorporates historical behavioral signals into reward design. By storing and leveraging intermediate model representations, we capture features of past rollouts and use density-based clustering to identify frequently recurring error patterns. Rollouts assigned to more prevalent error clusters are penalized more heavily, encouraging broader exploration while reducing repeated mistakes. Across five datasets and three base models, MEDS consistently improves average performance over existing baselines, achieving gains of up to 4.13 pass@1 points and 4.37 pass@128 points. Additional analyses using both LLM-based annotations and quantitative diversity metrics show that MEDS increases behavioral diversity during sampling.

Yang Liu, Enxi Wang, Yufei Gao +6

Fudan University · Shanghai Innovation Institute

RL for Language Models Reinforcement Learning Reward Design for RL Behavioral Diversity in RL

Apr 12, 2026

PokeRL: Reinforcement Learning for Pokemon Red

Pokemon Red is a long-horizon JRPG with sparse rewards, partial observability, and quirky control mechanics that make it a challenging benchmark for reinforcement learning. While recent work has shown that PPO agents can clear the first two gyms using heavy reward shaping and engineered observations, training remains brittle in practice, with agents often degenerating into action loops, menu spam, or unproductive wandering. In this paper, we present PokeRL, a modular system that trains deep reinforcement learning agents to complete early game tasks in Pokemon Red, including exiting the player's house, exploring Pallet Town to reach tall grass, and winning the first rival battle. Our main contributions are a loop-aware environment wrapper around the PyBoy emulator with map masking, a multi-layer anti-loop and anti-spam mechanism, and a dense hierarchical reward design. We argue that practical systems like PokeRL, which explicitly model failure modes such as loops and spam, are a necessary intermediate step between toy benchmarks and full Pokemon League champion agents. Code is available at https://github.com/reddheeraj/PokemonRL

Dheeraj Mudireddy, Sai Patibandla

Texas A&M Institute of Data Science Texas A&M University College Station, USA · Dept. of Computer Science Texas A&M University College Station, USA

RL for Game AI Reinforcement Learning RL Benchmarks Reward Design for RL Dense Reward RL

Apr 12, 2026

Learning Preference-Based Objectives from Clinical Narratives for Dynamic Sepsis Treatment

Designing reward functions for reinforcement learning (RL) in healthcare remains challenging because clinically meaningful outcomes are sparse, delayed, and difficult to explicitly specify. Although structured clinical data capture physiologic states, they often fail to reflect broader aspects of patient trajectories such as treatment response, recovery dynamics, and intervention burden. Clinical narratives, by contrast, encode longitudinal clinician assessments of disease progression, treatment effectiveness, and recovery, providing a potential source of trajectory-level supervision beyond predefined outcome metrics. We propose Clinical Narrative-informed Preference Rewards (CN-PR), a framework that learns reward functions directly from discharge summaries by treating clinical narratives as scalable supervision for trajectory-level preferences. Using a large language model, we derive trajectory quality scores and construct pairwise preferences between patient trajectories to learn rewards through preference-based optimization. To account for variability in narrative informativeness, we incorporate a task relevance signal that weights supervision according to its relevance to the downstream decision-making task. We evaluate CN-PR in dynamic sepsis treatment using offline RL. The learned reward demonstrated strong monotonic alignment with trajectory quality scores and produced policies associated with improved recovery-related outcomes, including increased organ support-free days and faster shock resolution, while maintaining mortality performance comparable to outcome-based reward baselines. These findings were preserved under external validation. Our results suggest that clinical narratives provide a scalable and expressive source of supervision for reward learning in dynamic treatment regimes.

Daniel J. Tan, Jayne Hui Zhen Chan, Kai Wen Hwang +3

Institute of Data Science, National University of Singapore, Singapore · National University Hospital, Singapore · Saw Swee Hock School of Public Health, National University of Singapore, Singapore

Reward Design for RL Reinforcement Learning Healthcare Preference-Based RL

Apr 9, 2026

On Tackling Complex Tasks with Reward Machines and Signal Temporal Logics

We propose a Reinforcement Learning (RL) based control design framework for handling complex tasks. The approach extends the concept of Reward Machines (RM) with Signal Temporal Logic (STL) formulas that can be used for event generation. The use of STL allows not only a more efficient representation of rewards for complex tasks but also guiding the training process to converge towards behaviors satisfying specified requirements. We also propose an implementation of the framework that leverages the STL online monitoring algorithms. We illustrate the framework with three case studies (minigrid, cart-pole and high-way environments) with non-trivial tasks.

Ana María Gómez Ruiz, Thao Dang, Alexandre Donzé

UGA · VERIMAG, Université Grenoble Alpes · VERIMAG - IMAG, CNRS, UGA +1

RL Control Reinforcement Learning Reward Design for RL

Apr 9, 2026

SPARD: Self-Paced Curriculum for RL Alignment via Integrating Reward Dynamics and Data Utility

The evolution of Large Language Models (LLMs) is shifting the focus from single, verifiable tasks toward complex, open-ended real-world scenarios, imposing significant challenges on the post-training phase. In these settings, the scale and complexity of reward systems have grown significantly, transitioning toward multi-objective formulations that encompass a comprehensive spectrum of model capabilities and application contexts. However, traditional methods typically rely on fixed reward weights, ignoring non-stationary learning dynamics and struggling with data heterogeneity across dimensions. To address these issues, we propose SPARD, a framework that establishes an automated, self-paced curriculum by perceiving learning progress to dynamically adjust multi-objective reward weights and data importance, thereby synchronizing learning intent with data utility for optimal performance. Extensive experiments across multiple benchmarks demonstrate that SPARD significantly enhances model capabilities across all domains.

Xuyang Zhi, Peilun zhou, Chengqiang Lu +10

University of Science and Technology of China · Xiaohongshu Inc.

RL Post-Training Multi-Objective Language Model Alignment Curriculum RL Reward Design for RL

Apr 5, 2026

Pedagogical Safety in Educational Reinforcement Learning: Formalizing and Detecting Reward Hacking in AI Tutoring Systems

Reinforcement learning (RL) is increasingly used to personalize instruction in intelligent tutoring systems, yet the field lacks a formal framework for defining and evaluating pedagogical safety. We introduce a four-layer model of pedagogical safety for educational RL comprising structural, progress, behavioral, and alignment safety and propose the Reward Hacking Severity Index (RHSI) to quantify misalignment between proxy rewards and genuine learning. We evaluate the framework in a controlled simulation of an AI tutoring environment with 120 sessions across four conditions and three learner profiles, totaling 18{,}000 interactions. Results show that an engagement-optimized agent systematically over-selected a high-engagement action with no direct mastery gain, producing strong measured performance but limited learning progress. A multi-objective reward formulation reduced this problem but did not eliminate it, as the agent continued to favor proxy-rewarding behavior in many states. In contrast, a constrained architecture combining prerequisite enforcement and minimum cognitive demand substantially reduced reward hacking, lowering RHSI from 0.317 in the unconstrained multi-objective condition to 0.102. Ablation results further suggest that behavioral safety was the most influential safeguard against repetitive low-value action selection. These findings suggest that reward design alone may be insufficient to ensure pedagogically aligned behavior in educational RL, at least in the simulated environment studied here. More broadly, the paper positions pedagogical safety as an important research problem at the intersection of AI safety and intelligent educational systems.

Oluseyi Olukola, Nick Rahimi

School of Computing Sciences and Computer Engineering, Hattiesburg, MS, USA

Reward Hacking Safe RL AI in Education Reward Design for RL

Apr 4, 2026

When Adaptive Rewards Hurt: Causal Probing and the Switching-Stability Dilemma in LLM-Guided LEO Satellite Scheduling

Adaptive reward design for deep reinforcement learning (DRL) in multi-beam LEO satellite scheduling is motivated by the intuition that regime-aware reward weights should outperform static ones. We systematically test this intuition and uncover a switching-stability dilemma: near-constant reward weights (342.1 Mbps) outperform carefully-tuned dynamic weights (103.3+/-96.8 Mbps) because PPO requires a quasistationary reward signal for value function convergence. Weight adaptation-regardless of quality-degrades performance by repeatedly restarting convergence. To understand why specific weights matter, we introduce a single-variable causal probing method that independently perturbs each reward term by +/-20% and measures PPO response after 50k steps. Probing reveals counterintuitive leverage: a +20% increase in the switching penalty yields +157 Mbps for polar handover and +130 Mbps for hot-cold regimes-findings inaccessible to human experts or trained MLPs without systematic probing. We evaluate four MDP architect variants (fixed, rule-based, learned MLP, finetuned LLM) across known and novel traffic regimes. The MLP achieves 357.9 Mbps on known regimes and 325.2 Mbps on novel regimes, while the fine-tuned LLM collapses to 45.3+/-43.0 Mbps due to weight oscillation rather than lack of domain knowledge-output consistency, not knowledge, is the binding constraint. Our findings provide an empirically-grounded roadmap for LLM-DRL integration in communication systems, identifying where LLMs add irreplaceable value (natural language intent understanding) versus where simpler methods suffice.

Yuanhang Li

Reward Design for RL LLM-Guided RL LEO Satellite Networks Non-Stationary RL

Apr 3, 2026

Multi-Turn Reinforcement Learning for Tool-Calling Agents with Iterative Reward Calibration

Training tool-calling agents with reinforcement learning on multi-turn tasks remains challenging due to sparse outcome rewards and difficult credit assignment across conversation turns. We present the first application of MT-GRPO (Multi-Turn Group Relative Policy Optimization) combined with GTPO (Generalized Token-level Policy Optimization) for training a tool-calling agent on realistic customer service tasks with an LLM-based user simulator. Through systematic analysis of training rollouts, we discover that naively designed dense per-turn rewards degrade performance by up to 14 percentage points due to misalignment between reward discriminativeness and advantage direction. We introduce Iterative Reward Calibration, a methodology for designing per-turn rewards using empirical discriminative analysis of rollout data, and show that our GTPO hybrid advantage formulation eliminates the advantage misalignment problem. Applied to the Tau-Bench airline benchmark, our approach improves Qwen3.5-4B from 63.8 percent to 66.7 percent (+2.9pp) and Qwen3-30B-A3B from 58.0 percent to 69.5 percent (+11.5pp) -- with the trained 4B model exceeding GPT-4.1 (49.4 percent) and GPT-4o (42.8 percent) despite being 50 times smaller, and the 30.5B MoE model approaching Claude Sonnet 4.5 (70.0 percent). To our knowledge, these are the first published RL training results on Tau-Bench. We release our code, reward calibration analysis, and training recipes.

Wachiravit Modecrua, Krittanon Kaewtawee, Krittin Pachtrachai +1

Amity Research and Application Center (ARAC)

Multi-Turn RL RL for Tool Use Reinforcement Learning Reward Design for RL Step-Level Credit Assignment in RL Group Relative Policy Optimization

Apr 3, 2026

Rubrics to Tokens: Bridging Response-level Rubrics and Token-level Rewards in Instruction Following Tasks

Rubric-based Reinforcement Learning (RL) has emerged as a promising approach for aligning Large Language Models (LLMs) with complex, open-domain instruction following tasks. However, existing methods predominantly rely on response-level rewards, introducing severe reward sparsity and reward ambiguity problems. To address these issues, we propose Rubrics to Tokens (RTT), a novel rubric-based RL framework that bridges coarse response-level scores and fine-grained token-level credit assignment. RTT introduces a Token-Level Relevance Discriminator to predict which tokens in the response are responsible for a specific constraint, and optimizes the policy model via RTT-GRPO, which integrates response-level and token-level advantages within a unified framework. Furthermore, when transitioning from one-dimensional, outcome-level reward to three-dimensional reward space in the token-level rubric-based RL, we propose a novel group normalization method, called Intra-sample Token Group Normalization, to accommodate this shift. Extensive experiments and benchmarks demonstrate that RTT consistently outperforms other baselines in both instruction- and rubric-level accuracy across different models.

Tianze Xu, Yanzhao Zheng, Pengrui Lu +11

Shanghai Jiao Tong University · GAIR · Alibaba Group +2

RL for Language Models Rubric-Based RL Instruction Following Reward Design for RL Step-Level Credit Assignment in RL

Apr 1, 2026

From Rebound to Remedy: Understanding and Mitigating Reward Hacking via Representation Engineering

Reinforcement learning for LLMs is vulnerable to reward hacking, where models exploit shortcuts to maximize reward without solving the intended task. We systematically study this phenomenon in coding tasks using an environment-manipulation setting, where models can rewrite evaluator code to trivially pass tests without solving the task, as a controlled testbed. Across both studied models, we identify a reproducible three-phase rebound pattern: models first attempt to rewrite the evaluator but fail, as their rewrites embed test cases their own solutions cannot pass. They then temporarily retreat to legitimate solving. When legitimate reward remains scarce, they rebound into successful hacking with qualitatively different strategies. Using representation engineering, we extract concept directions for shortcut, deception, and evaluation awareness from domain-general contrastive pairs and find that the shortcut direction tracks hacking behavior most closely, making it an effective representational proxy for detection. Motivated by this finding, we propose Advantage Modification, which integrates shortcut concept scores into GRPO advantage computation to penalize hacking rollouts before policy updates. Because the penalty is internalized into the training signal rather than applied only at inference time, Advantage Modification provides more robust suppression of hacking compared with generation-time activation steering.

Rui Wu, Ruixiang Tang

Department of Computer Science Rutgers University

Reward Hacking RL Fine-Tuning Reward Design for RL RL for Code Generation

Mar 31, 2026

Extending MONA in Camera Dropbox: Reproduction, Learned Approval, and Design Implications for Reward-Hacking Mitigation

Myopic Optimization with Non-myopic Approval (MONA) mitigates multi-step reward hacking by restricting the agent's planning horizon while supplying far-sighted approval as a training signal~\cite{farquhar2025mona}. The original paper identifies a critical open question: how the method of constructing approval -- particularly the degree to which approval depends on achieved outcomes -- affects whether MONA's safety guarantees hold. We present a reproduction-first extension of the public MONA Camera Dropbox environment that (i)~repackages the released codebase as a standard Python project with scripted PPO training, (ii)~confirms the published contrast between ordinary RL (91.5% reward-hacking rate) and oracle MONA (0.0% hacking rate) using the released reference arrays, and (iii)~introduces a modular learned-approval suite spanning oracle, noisy, misspecified, learned, and calibrated approval mechanisms. In reduced-budget pilot sweeps across approval methods, horizons, dataset sizes, and calibration strategies, the best calibrated learned-overseer run achieves zero observed reward hacking but substantially lower intended-behavior rates than oracle MONA (11.9% vs.\ 99.9%), consistent with under-optimization rather than re-emergent hacking. These results operationalize the MONA paper's approval-spectrum conjecture as a runnable experimental object and suggest that the central engineering challenge shifts from proving MONA's concept to building learned approval models that preserve sufficient foresight without reopening reward-hacking channels. Code, configurations, and reproduction commands are publicly available. https://github.com/codernate92/mona-camera-dropbox-repro

Nathan Heath

Independent Researcher

Reward Hacking Reward Modeling Reward Design for RL

Mar 31, 2026

ShapE-GRPO: Shapley-Enhanced Reward Allocation for Multi-Candidate LLM Training

In user-agent interaction scenarios such as recommendation, brainstorming, and code suggestion, Large Language Models (LLMs) often generate sets of candidate recommendations where the objective is to maximize the collective utility of the entire set rather than individual candidates independently. However, existing reinforcement learning post-training paradigms, such as Group Relative Policy Optimization (GRPO), typically assign the same set-level scalar reward to every candidate in the set. This leads to noisy training signals where poor candidates free-ride on the high reward produced by a single strong peer, resulting in suboptimal exploration. To address this, we propose Shapley-Enhanced GRPO (ShapE-GRPO). By leveraging the permutation-invariant nature of set-level utility, we derive a Shapley-enhanced formulation from cooperative game theory to decompose set-level rewards into granular, candidate-specific signals. We show that our formulation preserves the fundamental axioms of the Shapley value while remaining computationally efficient with polynomial-time complexity. Empirically, ShapE-GRPO consistently outperforms standard GRPO across diverse datasets with accelerated convergence during training.

Rui Ai, Yu Pan, David Simchi-Levi +1

MIT · University of Sydney · University of Texas at Dallas

RL Post-Training Reward Design for RL Credit Assignment in RL Group Relative Policy Optimization Shapley Value Attribution

Mar 31, 2026

Generalizable Dense Reward for Long-Horizon Robotic Tasks

Existing robotic foundation policies are trained primarily via large-scale imitation learning. While such models demonstrate strong capabilities, they often struggle with long-horizon tasks due to distribution shift and error accumulation. While reinforcement learning (RL) can finetune these models, it cannot work well across diverse tasks without manual reward engineering. We propose VLLR, a dense reward framework combining (1) an extrinsic reward from Large Language Models (LLMs) and Vision-Language Models (VLMs) for task progress recognition, and (2) an intrinsic reward based on policy self-certainty. VLLR uses LLMs to decompose tasks into verifiable subtasks and then VLMs to estimate progress to initialize the value function for a brief warm-up phase, avoiding prohibitive inference cost during full training; and self-certainty provides per-step intrinsic guidance throughout PPO finetuning. Ablation studies reveal complementary benefits: VLM-based value initialization primarily improves task completion efficiency, while self-certainty primarily enhances success rates, particularly on out-of-distribution tasks. On the CHORES benchmark covering mobile manipulation and navigation, VLLR achieves up to 56% absolute success rate gains over the pretrained policy, up to 5% gains over state-of-the-art RL finetuning methods on in-distribution tasks, and up to

10\%

gains on out-of-distribution tasks, all without manual reward engineering. Additional visualizations can be found in https://silongyong.github.io/vllr_project_page/

Silong Yong, Stephen Sheng, Carl Qi +6

Carnegie Mellon University · Amazon Robotics · UT Austin

Reinforcement Learning Long-Horizon Robotic Manipulation Robot Skill Learning Dense Reward RL Intrinsic Motivation LLM-Guided RL+3

Mar 25, 2026

Large Language Model Guided Incentive Aware Reward Design for Cooperative Multi-Agent Reinforcement Learning

Designing effective auxiliary rewards for cooperative multi-agent systems remains challenging, as misaligned incentives can induce suboptimal coordination, particularly when sparse task rewards provide insufficient grounding for coordinated behavior. This study introduces an autonomous reward design framework that uses large language models (LLMs) to synthesize executable reward programs from environment instrumentation. The procedure constrains candidate programs within a formal validity envelope and trains policies from scratch using Multi-Agent Proximal Policy Optimization (MAPPO) under a fixed computational budget. The candidates are then evaluated on the basis of their performance, and selection across generations solely based on the sparse task returns. The framework is evaluated in four Overcooked-AI layouts characterized by varying levels of corridor congestion, handoff dependencies, and structural asymmetries. The proposed reward design approach consistently yields higher task returns and delivery counts, with the most pronounced gains observed in environments dominated by interaction bottlenecks. Diagnostic analysis of the synthesized shaping components reveals stronger interdependence in action selection and improved signal alignment in coordination-intensive tasks. These results demonstrate that the proposed LLM-guided reward search framework mitigates the need for manual engineering while producing shaping signals compatible with cooperative learning under finite budgets.

Dogan Urgun, Gokhan Gungor

Department of Electrical and Electronics Engineering, Karabuk University, 78050 Karabuk, Türkiye · Department of Mechatronics Engineering, Karabuk University, 78050 Karabuk, Türkiye

Reward Design for RL Cooperative MARL Multi-Agent Reinforcement Learning LLM-Guided RL

Mar 20, 2026

Experience is the Best Teacher: Motivating Effective Exploration in Reinforcement Learning for LLMs

Reinforcement Learning (RL) with rubric-based rewards has recently shown remarkable progress in enhancing general reasoning capabilities of Large Language Models (LLMs), yet still suffers from ineffective exploration confined to curent policy distribution. In fact, RL optimization can be viewed as steering the policy toward an ideal distribution that maximizes the rewards, while effective exploration should align efforts with desired target. Leveraging this insight, we propose HeRL, a Hindsight experience guided Reinforcement Learning framework to bootstrap effective exploration by explicitly telling LLMs the desired behaviors specified in rewards. Concretely, HeRL treats failed trajectories along with their unmet rubrics as hindsight experience, which serves as in-context guidance for the policy to explore desired responses beyond its current distribution. Additionally, we introduce a bonus reward to incentivize responses with greater potential for improvement under such guidance. HeRL facilitates effective learning from desired high quality samples without repeated trial-and-error from scratch, yielding a more accurate estimation of the expected gradient theoretically. Extensive experiments across various benchmarks demonstrate that HeRL achieves superior performance gains over baselines, and can further benefit from experience guided self-improvement at test time. Our code is available at https://github.com/sikelifei/HeRL.

Wenjian Zhang, Kongcheng Zhang, Jiaxin Qi +2

Dalian University of Technology · Zhejiang University · Chinese Academy of Sciences

Hindsight Experience Replay RL for Language Model Reasoning RL Exploration Rubric-Based RL In-Context RL Reward Design for RL

Mar 20, 2026

Decomposable Reward Modeling and Realistic Environment Design for Reinforcement Learning-Based Forex Trading

Applying reinforcement learning (RL) to foreign exchange (Forex) trading remains challenging because realistic environments, well-defined reward functions, and expressive action spaces must be satisfied simultaneously, yet many prior studies rely on simplified simulators, single scalar rewards, and restricted action representations, limiting both interpretability and practical relevance. This paper presents a modular RL framework designed to address these limitations through three tightly integrated components: a friction-aware execution engine that enforces strict anti-lookahead semantics, with observations at time t, execution at time t+1, and mark-to-market at time t+1, while incorporating realistic costs such as spread, commission, slippage, rollover financing, and margin-triggered liquidation; a decomposable 11-component reward architecture with fixed weights and per-step diagnostic logging to enable systematic ablation and component-level attribution; and a 10-action discrete interface with legal-action masking that encodes explicit trading primitives while enforcing margin-aware feasibility constraints. Empirical evaluation on EURUSD focuses on learning dynamics rather than generalization and reveals strongly non-monotonic reward interactions, where additional penalties do not reliably improve outcomes; the full reward configuration achieves the highest training Sharpe (0.765) and cumulative return (57.09 percent). The expanded action space increases return but also turnover and reduces Sharpe relative to a conservative 3-action baseline, indicating a return-activity trade-off under a fixed training budget, while scaling-enabled variants consistently reduce drawdown, with the combined configuration achieving the strongest endpoint performance.

Nabeel Ahmad Saidd

Dr APJ Abdul Kalam Technical University, India

Reinforcement Learning Reward Design for RL Finance

Mar 19, 2026

RewardFlow: Topology-Aware Reward Propagation on State Graphs for Agentic RL with Large Language Models

Reinforcement learning (RL) shows promise for enhancing LLM agentic reasoning, yet sparse terminal rewards hinder fine-grained optimization. Process reward modeling offers an alternative but incurs high computational costs, reward hacking risks, and annotation bottlenecks. We introduce RewardFlow, a lightweight method for estimating state-level rewards in agentic reasoning. By constructing state graphs that capture the intrinsic topological structure of trajectories, RewardFlow performs topology-aware propagation to estimate each state's contribution to success, yielding principled, annotation-free dense rewards. Used for RL optimization, RewardFlow substantially outperforms prior baselines across four agentic benchmarks: +6.2% average success rate on text-based tasks, +29.7% on visual reasoning over the strongest baseline across three model scales, and +10% accuracy on DeepResearch, with superior robustness and training efficiency. The implementation of RewardFlow is publicly available at https://github.com/tmlr-group/RewardFlow.

Xiao Feng, Bo Han, Zhanke Zhou +5

TMLR Group, Hong Kong Baptist University · TCL Corporate Research (HK) Co., Ltd · Cooperative Medianet Innovation Center, Shanghai Jiao Tong University +1

RL for LLM Agents Reward Design for RL Dense Reward RL Credit Assignment in RL

Mar 18, 2026

Escaping Offline Pessimism: Vector-Field Reward Shaping for Safe Frontier Exploration

While offline reinforcement learning provides reliable policies for real-world deployment, its inherent pessimism severely restricts an agent's ability to explore and collect novel data online. Drawing inspiration from safe reinforcement learning, exploring near the boundary of regions well covered by the offline dataset and reliably modeled by the simulator allows an agent to take manageable risks--venturing into informative but moderate-uncertainty states while remaining close enough to familiar regions for safe recovery. However, naively rewarding this boundary-seeking behavior can lead to a degenerate parking behavior, where the agent simply stops once it reaches the frontier. To solve this, we propose a novel vector-field reward shaping paradigm designed to induce continuous, safe boundary exploration for non-adaptive deployed policies. Operating on an uncertainty oracle trained from offline data, our reward combines two complementary components: a gradient-alignment term that attracts the agent toward a target uncertainty level, and a rotational-flow term that promotes motion along the local tangent plane of the uncertainty manifold. Through theoretical analysis, we show that this reward structure naturally induces sustained exploratory behavior along the boundary while preventing degenerate solutions. Empirically, by integrating our proposed reward shaping with Soft Actor-Critic on a 2D continuous navigation task, we validate that agents successfully traverse uncertainty boundaries while balancing safe, informative data collection with primary task completion.

Amirhossein Roknilamouki, Arnob Ghosh, Eylem Ekici +1

Department of Electrical and Computer Engineering, The Ohio State University · Department of Electrical and Computer Engineering, New Jersey Institute of Technology · Department of Computer Science and Engineering, The Ohio State University.

RL Exploration Reward Design for RL Frontier-Based Exploration Offline-To-Online RL Safe RL

Mar 17, 2026

Rewarding DINO: Predicting Dense Rewards with Vision Foundation Models

Well-designed dense reward functions in robot manipulation not only indicate whether a task is completed but also encode progress along the way. Generally, designing dense rewards is challenging and usually requires access to privileged state information available only in simulation, not in real-world experiments. This makes reward prediction models that infer task state information from camera images attractive. A common approach is to predict rewards from expert demonstrations based on visual similarity or sequential frame ordering. However, this biases the resulting reward function towards a specific solution and leaves it undefined in states not covered by the demonstrations. In this work, we introduce Rewarding DINO, a method for language-conditioned reward modeling that learns actual reward functions rather than specific trajectories. The model's compact size allows it to serve as a direct replacement for analytical reward functions with comparatively low computational overhead. We train our model on data sampled from 24 Meta-World+ tasks using a rank-based loss and evaluate pairwise accuracy, rank correlation, and calibration. Rewarding DINO achieves competitive performance in tasks from the training set and generalizes to new settings in simulation and the real world, indicating that it learns task semantics. We also test the model with off-the-shelf reinforcement learning algorithms to solve tasks from our Meta-World+ training set.

Pierre Krack, Tobias Jülg, Wolfram Burgard +1

Department of Computer Science & Artificial Intelligence, University of Technology Nuremberg, Germany. · TUM School of Computation, Information and Technology, Technical University of Munich, Germany

Reward Modeling Reward Design for RL Dense Reward RL Multimodal Reward Modeling

Mar 17, 2026

From the Inside Out: Progressive Distribution Refinement for Confidence Calibration

Leveraging the model's internal information as the self-reward signal in Reinforcement Learning (RL) has received extensive attention due to its label-free nature. While prior works have made significant progress in applying the Test-Time Scaling (TTS) strategies to RL, the discrepancy in internal information between test and training remains inadequately addressed. Moreover, Test-Time Training based on voting-based TTS strategies often suffers from reward hacking problems. To address these issues, we propose DistriTTRL, which leverages the distribution prior of the model's confidence during RL to progressively optimize the reward signal, rather than relying solely on single-query rollouts. Additionally, we mitigate the phenomenon of consistent reward hacking caused by the voting-based TTS strategies through diversity-targeted penalties. Benefiting from this training mechanism where model capability and self-reward signals complement each other, and the mitigation of reward hacking, DistriTTRL has achieved significant performance improvements across multiple models and benchmarks.

Xizhong Yang, Yinan Xia, Huiming Wang +1

Southeast University · Kuaishou Technology · SUTD

Test-Time RL Reinforcement Learning Reward Design for RL Probability Calibration Reward Hacking

Mar 17, 2026

Large Reward Models: Generalizable Online Robot Reward Generation with Vision-Language Models

Reinforcement Learning (RL) has shown strong potential for improving robotic manipulation policies, yet its practical use remains bottlenecked by the difficulty of specifying reward functions that are both semantically meaningful and reusable across tasks. In this paper, we propose Large Reward Models (LRMs), a framework that adapts foundation VLMs into frame-level reward generators for robot policy refinement. We specialize a state-of-the-art VLM on a multi-source dataset spanning real-world robot trajectories, human-object interactions, and simulated manipulation environments. Unlike prior approaches that mainly evaluate trajectories post-hoc, LRMs expose multiple reward interfaces from visual observations: progress estimation, task completion, and temporal contrastive comparison. Starting from an imitation-learned policy, we use these VLM-derived rewards to guide PPO refinement on held-out long-horizon manipulation tasks. Our experiments show that LRM progress rewards provide the strongest non-privileged online refinement signal, improving the IL baseline and narrowing the gap to privileged environment rewards. We further deploy progress rewards for progress-weighted behavioral cloning on four real-world manipulation tasks spanning two robot platforms, improving over SFT on all four tasks. These results suggest that modality-specific specialization of foundation VLMs can provide practical visual reward signals for both simulated policy refinement and physical robot self-improvement without hand-coded task rewards.

Yanru Wu, Weiduo Yuan, Esteban Martinez Licon +4

USC Physical Superintelligence Lab

Long-Horizon Robotic Manipulation Robot Skill Learning Robotic RL Multimodal Reward Modeling Visual RL VLMs for Robotics+1