# Generalizable Dense Reward for Long-Horizon Robotic Tasks

arXiv: https://arxiv.org/abs/2604.00055 · HTML: https://mlsanity.com/paper/2604.00055/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in Language-conditioned policies and VLAs, one of 13 directions in Offline RL and imitation learning from data.
- Builds on 32 of its 39 references that are in the map.
- First posted on arXiv 31 Mar 2026.

## Placed in

- Offline RL and imitation learning from data / [Language-conditioned policies and VLAs](https://mlsanity.com/offlinerl/language-conditioned-policies-and-vlas/)

## Builds on (most cited first, 30 of 32)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor](https://mlsanity.com/paper/1801.01290/) (2018)
- [Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2204.05862/) (2022)
- [Playing Atari with Deep Reinforcement Learning](https://mlsanity.com/paper/1312.5602/) (2013)
- [RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control](https://mlsanity.com/paper/2307.15818/) (2023)
- [OpenVLA: An Open-Source Vision-Language-Action Model](https://mlsanity.com/paper/2406.09246/) (2024)
- [Trust Region Policy Optimization](https://mlsanity.com/paper/1502.05477/) (2015)
- [Addressing Function Approximation Error in Actor-Critic Methods](https://mlsanity.com/paper/1802.09477/) (2018)
- [π0: A Vision-Language-Action Flow Model for General Robot Control](https://mlsanity.com/paper/2410.24164/) (2024)
- [RT-1: Robotics Transformer for Real-World Control at Scale](https://mlsanity.com/paper/2212.06817/) (2022)
- [GR00T N1: An Open Foundation Model for Generalist Humanoid Robots](https://mlsanity.com/paper/2503.14734/) (2025)
- [Eureka: Human-Level Reward Design via Coding Large Language Models](https://mlsanity.com/paper/2310.12931/) (2023)
- [ProcTHOR: Large-Scale Embodied AI Using Procedural Generation](https://mlsanity.com/paper/2206.06994/) (2022)
- [An Embodied Generalist Agent in 3D World](https://mlsanity.com/paper/2311.12871/) (2023)
- [Learning to Reason without External Rewards](https://mlsanity.com/paper/2505.19590/) (2025)
- [Scalable Best-of-N Selection for Large Language Models via Self-Certainty](https://mlsanity.com/paper/2502.18581/) (2025)
- [VLA-RL: Towards Masterful and General Robotic Manipulation with Scalable Reinforcement Learning](https://mlsanity.com/paper/2505.18719/) (2025)
- [Vision-Language Models are Zero-Shot Reward Models for Reinforcement Learning](https://mlsanity.com/paper/2310.12921/) (2023)
- [Jump-Start Reinforcement Learning](https://mlsanity.com/paper/2204.02372/) (2022)
- [GRAPE: Generalizing Robot Policy via Preference Alignment](https://mlsanity.com/paper/2411.19309/) (2024)
- [PIRLNav: Pretraining with Imitation and RL Finetuning for ObjectNav](https://mlsanity.com/paper/2301.07302/) (2023)
- [Vision Language Models are In-Context Value Learners](https://mlsanity.com/paper/2411.04549/) (2024)
- [FLaRe: Achieving Masterful and Adaptive Robot Policies with Large-Scale Reinforcement Learning Fine-Tuning](https://mlsanity.com/paper/2409.16578/) (2024)
- [SPOC: Imitating Shortest Paths in Simulation Enables Effective Navigation and Manipulation in the Real World](https://mlsanity.com/paper/2312.02976/) (2023)
- [Long-VLA: Unleashing Long-Horizon Capability of Vision Language Action Model for Robot Manipulation](https://mlsanity.com/paper/2508.19958/) (2025)
- [Selective Visual Representations Improve Convergence and Generalization for Embodied AI](https://mlsanity.com/paper/2311.04193/) (2023)
- [General agents contain world models](https://mlsanity.com/paper/2506.01622/) (2025)

## Built on by (most cited first, 3 of 3)

- [ProcVLM: Learning Procedure-Grounded Progress Rewards for Robotic Manipulation](https://mlsanity.com/paper/2605.08774/) (2026)
- [Progress Reward Modeling for Robotic Learning: A Comprehensive Survey](https://mlsanity.com/paper/2607.21655/) (2026)
- [Foresight: Iterative Reasoning About Clues that Matter for Navigation](https://mlsanity.com/paper/2606.12550/) (2026)

## Abstract (arXiv)

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/vllrprojectpage/