mlsanityData to 14 Oct 2026

Intrinsic Robot Rewarding: Reusing VLA Representations for Autonomous Evaluation and Policy Improvement

arXiv 2609.17115 ↗First posted 15 Sep 2026cs.RO cs.LGcited by 0

On the map: Vision-language-action models → Reinforcement learning and post-training

Abstract (arXiv)

Vision-language-action (VLA) systems already bring together two valuable resources for robot learning: rich visual representations and demonstrations of successful task execution. Intrinsic Robot Rewarding (IRR) proposes to use these resources for a second, complementary purpose: evaluating the robot's own outcomes and providing feedback for policy improvement. Successful demonstration endpoints define task-specific references, and the policy's frozen visual encoder provides the feature space in which new outcomes are assessed. The core reward mechanism adds a reference bank and a scoring operation to the existing pipeline, without requiring a separate learned evaluator or an additional perception backbone. Our position is that this reuse offers a promising route to lower integration effort, efficient reward computation, and reduced recurring human outcome scoring. Building on established research in visual rewards and learning from experience, IRR brings these ideas into the robot's existing perception and demonstration pipeline. An operational COMAU Racer 3 demonstrator is available at technology readiness level 4 (TRL 4). This laboratory foundation supports the next research step: connecting internal outcome evaluation to physical policy improvement. We present the reward formulation, central research questions, and an evaluation methodology linking reward reliability to task success and supervision effort. The intended contribution is a reusable approach to learn and improve from the data and experience already available in industrial robot systems.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Reinforcement learning and post-training, one of 13 directions in Vision-language-action models. Builds on 20 of its 22 references that are in the map. First posted on arXiv 15 Sep 2026.

Builds on

20 of 22 references.

Closest ancestors · 2

Same direction, newest first

  1. 2025SRPO: Self-Referential Policy Optimization for Vision-Language-Action Models19 Nov
  2. 2025π*0.6: a VLA That Learns From Experience18 Nov

Other references · 18

  1. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,974
  2. 2018Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actorcited by 4,885
  3. 2023Sigmoid Loss for Language Image Pre-Trainingcited by 3,109
  4. 2023RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Controlcited by 3,080
  5. 2024OpenVLA: An Open-Source Vision-Language-Action Modelcited by 2,887
  6. 2017Curiosity-driven Exploration by Self-supervised Predictioncited by 1,194
  7. 2025Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Successcited by 883
  8. 2025V-JEPA 2: Self-Supervised Video Models Enable Understanding, Prediction and Planningcited by 809
  9. 2022VIP: Towards Universal Visual Reward and Representation via Value-Implicit Pre-Trainingcited by 395
  10. 2018Residual Reinforcement Learning for Robot Controlcited by 274
  11. 2024Precise and Dexterous Robotic Manipulation via Human-in-the-Loop Reinforcement Learningcited by 189
  12. 2023LIV: Language-Image Representations and Rewards for Robotic Controlcited by 188
6 more
  1. 2024RL-VLM-F: Reinforcement Learning from Vision Language Foundation Model Feedbackcited by 156
  2. 2023RoboCLIP: One Demonstration is Enough to Learn Robot Policiescited by 134
  3. 2023Vision-Language Models as a Source of Rewardscited by 66
  4. 2026RoboReward: General-Purpose Vision-Language Reward Models for Roboticscited by 53
  5. 2026Large Reward Models: Generalizable Online Robot Reward Generation with Vision-Language Modelscited by 11
  6. 2025GoalLadder: Incremental Goal Discovery with Vision-Language Modelscited by 3

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025A Vision-Language-Action-Critic Model for Robotic Real-World Reinforcement Learning
  2. 2026Generalizable Dense Reward for Long-Horizon Robotic Tasks
  3. 2026SARM2: Multi-Task Stage Aware Reward Modeling for Self Improving Robotic Manipulation
  4. 2026ARS: Agentic Reward System for Robot Learning
  5. 2025RoVer: Robot Reward Model as Test-Time Verifier for Vision-Language-Action Model
  6. 2025VARP: Reinforcement Learning from Vision-Language Model Feedback with Agent Regularized Preferences
  7. 2026From Demonstrations to Rewards: Test-Time Prompt Optimization for VLM Reward Models
  8. 2024Video-Language Critic: Transferable Reward Functions for Language-Conditioned Robotics