arXiv subjects

Erdem Biyik

Publications and source records attributed to Erdem Biyik.

14 recordsLinked to original sources

arXiv2026

From LLM-Generated Specifications to Learned Quadruped Locomotion

Quadruped robot locomotion policies are often trained using reinforcement learning, which in turn relies heavily on hand-crafted reward functions. Designing reward functions requires substantial manual engineering, and it is often unclear which local rewards will induce the desired global behavior. Shaped rewards from formal specifications in languages like Signal Temporal Logic (STL) can make rewards more interpretable, but writing STL specifications itself still requires domain expertise. We study whether large language models (LLMs) can fill this gap by generating Parametric Signal Temporal Logic (PSTL) specifications that are subsequently used for policy learning. Given a natural language locomotion objective and a constrained specification grammar, GPT-5.5 and Qwen 3.6 independently propose STL templates for command tracking, safety, and gait structure. We instantiate the parameters of the generated PSTL templates using expert trajectories and retain only specifications that are consistent with demonstrated expert behavior. The resulting specifications are then transformed into smooth, finite-history reward functions and used to train a quadruped locomotion policy with Proximal Policy Optimization (PPO) in MuJoCo XLA (MJX). We evaluate both \emph{gait-aware} and \emph{gait-agnostic} settings. The former specifies walking-trot, trot, and bound regimes, while the latter allows contact patterns to emerge from the task objective. We compare against hand-engineered rewards, Text2Reward-style LLM-generated reward code, and an expert-switching oracle. Gait-aware Qwen 3.6 specifications achieved 100\% survival and command success across all tested speeds (0.3--2.1 m/s) and matched the target gait at high speeds, whereas Text2Reward achieved 0\% for both metrics at $\geq 1.9$ m/s. Videos: https://stl-locomotion.github.io/

cs.RO↗

arXiv2026

Robometer: Scaling General-Purpose Robotic Reward Models via Trajectory Comparisons

General-purpose robot reward models are typically trained to predict absolute task progress from expert demonstrations, providing only local, frame-level supervision. While effective for expert demonstrations, this paradigm scales poorly to large-scale robotics datasets where failed and suboptimal trajectories are abundant and assigning dense progress labels is ambiguous. We introduce Robometer, a scalable reward modeling framework that combines intra-trajectory progress supervision with inter-trajectory preference supervision. Robometer is trained with a dual objective: a frame-level progress loss that anchors reward magnitude on expert data, and a trajectory-comparison preference loss that imposes global ordering constraints across trajectories of the same task, enabling effective learning from both real and augmented failed trajectories. To support this formulation at scale, we curate RBM-1M, a reward-learning dataset comprising over one million trajectories spanning diverse robot embodiments and tasks, including substantial suboptimal and failure data. Across benchmarks and real-world evaluations, Robometer learns more generalizable reward functions than prior methods and improves robot learning performance across a diverse set of downstream applications. Code, model weights, and videos at https://robometer.github.io/.

cs.RO↗

arXiv2026

Beyond Explainable AI (XAI): An Overdue Paradigm Shift and Post-XAI Research Directions

This study provides a cross-disciplinary examination of Explainable Artificial Intelligence (XAI) approaches-focusing on deep neural networks (DNNs) and large language models (LLMs)-and identifies empirical and conceptual limitations in current XAI. We discuss critical symptoms that stem from deeper root causes (i.e., two paradoxes, two conceptual confusions, and five false assumptions). These fundamental problems within the current XAI research field reveal three insights: experimentally, XAI exhibits significant flaws; conceptually, it is paradoxical; and pragmatically, further attempts to reform the paradoxical XAI might exacerbate its confusion-demanding fundamental shifts and new research directions. To move beyond XAI's limitations, we propose a four-pronged synthesized paradigm shift toward reliable and certified AI development. These four components include: verification-focused Interactive AI (IAI) to establish scientific community protocols for certifying AI system performance rather than attempting post-hoc explanations, AI Epistemology for rigorous scientific foundations, User-Sensible AI to create context-aware systems tailored to specific user communities, and Model-Centered Interpretability for faithful technical analysis-together offering comprehensive post-XAI research directions.

cs.CY↗

arXiv2026

SyncTwin: Fast Digital Twin Construction and Synchronization for Safe Robotic Manipulation

Accurate and safe robotic manipulation under dynamic and visually occluded conditions remains a core challenge in real-world deployment. We introduce SyncTwin, a novel digital twin framework that unifies fast 3D scene reconstruction and real-to-sim synchronization for robust and safety-aware robotic manipulation in such environments. In the offline stage, we employ VGGT to rapidly reconstruct object-level 3D assets from RGB images, forming a reusable geometry library. During execution, SyncTwin continuously synchronizes the digital twin by tracking real-world object states via point cloud segmentation updates and aligning them through colored-ICP registration. The synchronized twin enables motion planners to compute collision-free and dynamically feasible trajectories in simulation, which are safely executed on the real robot through a closed real-to-sim-to-real loop. Experiments in dynamic and occluded scenes show that SyncTwin improves manipulation performance and motion safety, demonstrating the effectiveness of digital twin synchronization for real-world robotic execution. The video demos and code can be found on the project website: https://sync-twin.github.io/.

cs.RO↗

arXiv2025

ReWiND: Language-Guided Rewards Teach Robot Policies without New Demonstrations

We introduce ReWiND, a framework for learning robot manipulation tasks solely from language instructions without per-task demonstrations. Standard reinforcement learning (RL) and imitation learning methods require expert supervision through human-designed reward functions or demonstrations for every new task. In contrast, ReWiND starts from a small demonstration dataset to learn: (1) a data-efficient, language-conditioned reward function that labels the dataset with rewards, and (2) a language-conditioned policy pre-trained with offline RL using these rewards. Given an unseen task variation, ReWiND fine-tunes the pre-trained policy using the learned reward function, requiring minimal online interaction. We show that ReWiND's reward model generalizes effectively to unseen tasks, outperforming baselines by up to 2.4x in reward generalization and policy alignment metrics. Finally, we demonstrate that ReWiND enables sample-efficient adaptation to new tasks, beating baselines by 2x in simulation and improving real-world pretrained bimanual policies by 5x, taking a step towards scalable, real-world robot learning. See website at https://rewind-reward.github.io/.

cs.RO↗

arXiv2025

CLAM: Continuous Latent Action Models for Robot Learning from Unlabeled Demonstrations

Learning robot control policies from demonstrations typically requires action-labeled expert data, which is expensive to collect through teleoperation. We study a more practical setting in which expert demonstrations are available only as observation sequences without action labels, and only task-agnostic play data contains actions. We introduce continuous latent action models (CLAM), a framework that infers continuous latent actions between consecutive observations using self-supervised dynamics prediction. To ground these latent actions into executable motor commands, CLAM jointly trains an action decoder using a small amount of task-agnostic play data. We show that continuous latent actions combined with this joint training are essential for high-dimensional continuous control. Across DMControl locomotion, MetaWorld manipulation, and real-world WidowX robot tasks, CLAM improves average task success rates by 2-3x over prior latent-action baselines and approaches behavior cloning trained with privileged expert action labels. Our results demonstrate that effective robot policies can be learned from unlabeled demonstrations and deployed on real hardware without collecting expert action-labeled data. Videos and code are available at clamrobot.github.io.

cs.RO↗

arXiv2024

EXTRACT: Efficient Policy Learning by Extracting Transferable Robot Skills from Offline Data

Most reinforcement learning (RL) methods focus on learning optimal policies over low-level action spaces. While these methods can perform well in their training environments, they lack the flexibility to transfer to new tasks. Instead, RL agents that can act over useful, temporally extended skills rather than low-level actions can learn new tasks more easily. Prior work in skill-based RL either requires expert supervision to define useful skills, which is hard to scale, or learns a skill-space from offline data with heuristics that limit the adaptability of the skills, making them difficult to transfer during downstream RL. Our approach, EXTRACT, instead utilizes pre-trained vision language models to extract a discrete set of semantically meaningful skills from offline data, each of which is parameterized by continuous arguments, without human supervision. This skill parameterization allows robots to learn new tasks by only needing to learn when to select a specific skill and how to modify its arguments for the specific task. We demonstrate through experiments in sparse-reward, image-based, robot manipulation environments that EXTRACT can more quickly learn new tasks than prior works, with major gains in sample efficiency and performance over prior skill-based RL. Website at https://www.jessezhang.net/projects/extract/.

cs.RO↗

arXiv2024

RL-VLM-F: Reinforcement Learning from Vision Language Foundation Model Feedback

Reward engineering has long been a challenge in Reinforcement Learning (RL) research, as it often requires extensive human effort and iterative processes of trial-and-error to design effective reward functions. In this paper, we propose RL-VLM-F, a method that automatically generates reward functions for agents to learn new tasks, using only a text description of the task goal and the agent's visual observations, by leveraging feedbacks from vision language foundation models (VLMs). The key to our approach is to query these models to give preferences over pairs of the agent's image observations based on the text description of the task goal, and then learn a reward function from the preference labels, rather than directly prompting these models to output a raw reward score, which can be noisy and inconsistent. We demonstrate that RL-VLM-F successfully produces effective rewards and policies across various domains - including classic control, as well as manipulation of rigid, articulated, and deformable objects - without the need for human supervision, outperforming prior methods that use large pretrained models for reward generation under the same assumptions. Videos can be found on our project website: https://rlvlmf2024.github.io/

cs.RO↗

arXiv2023

Value Explicit Pretraining for Learning Transferable Representations

Understanding visual inputs for a given task amidst varied changes is a key challenge posed by visual reinforcement learning agents. We propose \textit{Value Explicit Pretraining} (VEP), a method that learns generalizable representations for transfer reinforcement learning. VEP enables efficient learning of new tasks that share similar objectives as previously learned tasks, by learning an encoder that trains representations to be invariant to changes in environment dynamics and appearance. To pretrain the encoder with \textit{suboptimal unlabeled demonstration data} (sequence of observations and sparse reward signals), we use a self-supervised contrastive loss that enables the model to relate states across different tasks based on the Monte Carlo value estimate that is reflective of task progress, resulting in temporally smooth representations that capture the objective of the task. A major difference between our method and the existing approaches is the use of suboptimal unlabeled data that do not always solve the task. Experiments on Ant locomotion, a realistic navigation simulator and the Atari benchmark show that VEP outperforms current SoTA pretraining methods on the ability to generalize to unseen tasks. VEP achieves up to $2\times$ improvement in rewards, and up to $3\times$ improvement in sample efficiency. For videos of VEP policies, visit our \href{https://sites.google.com/view/value-explicit-pretraining/}{website}.

cs.LG↗

arXiv2023

Preference Elicitation with Soft Attributes in Interactive Recommendation

Preference elicitation plays a central role in interactive recommender systems. Most preference elicitation approaches use either item queries that ask users to select preferred items from a slate, or attribute queries that ask them to express their preferences for item characteristics. Unfortunately, users often wish to describe their preferences using soft attributes for which no ground-truth semantics is given. Leveraging concept activation vectors for soft attribute semantics, we develop novel preference elicitation methods that can accommodate soft attributes and bring together both item and attribute-based preference elicitation. Our techniques query users using both items and soft attributes to update the recommender system's belief about their preferences to improve recommendation quality. We demonstrate the effectiveness of our methods vis-a-vis competing approaches on both synthetic and real-world datasets.

cs.IR↗

arXiv2022

Assistive Teaching of Motor Control Tasks to Humans

Recent works on shared autonomy and assistive-AI technologies, such as assistive robot teleoperation, seek to model and help human users with limited ability in a fixed task. However, these approaches often fail to account for humans' ability to adapt and eventually learn how to execute a control task themselves. Furthermore, in applications where it may be desirable for a human to intervene, these methods may inhibit their ability to learn how to succeed with full self-control. In this paper, we focus on the problem of assistive teaching of motor control tasks such as parking a car or landing an aircraft. Despite their ubiquitous role in humans' daily activities and occupations, motor tasks are rarely taught in a uniform way due to their high complexity and variance. We propose an AI-assisted teaching algorithm that leverages skill discovery methods from reinforcement learning (RL) to (i) break down any motor control task into teachable skills, (ii) construct novel drill sequences, and (iii) individualize curricula to students with different capabilities. Through an extensive mix of synthetic and user studies on two motor control tasks -- parking a car with a joystick and writing characters from the Balinese alphabet -- we show that assisted teaching with skills improves student performance by around 40% compared to practicing full trajectories without skills, and practicing with individualized drills can result in up to 25% further improvement. Our source code is available at https://github.com/Stanford-ILIAD/teaching

cs.AI↗

arXiv2020

When Humans Aren't Optimal: Robots that Collaborate with Risk-Aware Humans

In order to collaborate safely and efficiently, robots need to anticipate how their human partners will behave. Some of today's robots model humans as if they were also robots, and assume users are always optimal. Other robots account for human limitations, and relax this assumption so that the human is noisily rational. Both of these models make sense when the human receives deterministic rewards: i.e., gaining either $100 or $130 with certainty. But in real world scenarios, rewards are rarely deterministic. Instead, we must make choices subject to risk and uncertainty--and in these settings, humans exhibit a cognitive bias towards suboptimal behavior. For example, when deciding between gaining $100 with certainty or $130 only 80% of the time, people tend to make the risk-averse choice--even though it leads to a lower expected gain! In this paper, we adopt a well-known Risk-Aware human model from behavioral economics called Cumulative Prospect Theory and enable robots to leverage this model during human-robot interaction (HRI). In our user studies, we offer supporting evidence that the Risk-Aware model more accurately predicts suboptimal human behavior. We find that this increased modeling accuracy results in safer and more efficient human-robot collaboration. Overall, we extend existing rational human models so that collaborative robots can anticipate and plan around suboptimal human behavior during HRI.

cs.RO↗

arXiv2017

Reconstruction by Calibration over Tensors for Multi-Coil Multi-Acquisition Balanced SSFP Imaging

Purpose: To develop a rapid imaging framework for balanced steady-state free precession (bSSFP) that jointly reconstructs undersampled data (by a factor of R) across multiple coils (D) and multiple acquisitions (N). To devise a multi-acquisition coil compression technique for improved computational efficiency. Methods: The bSSFP image for a given coil and acquisition is modeled to be modulated by a coil sensitivity and a bSSFP profile. The proposed reconstruction by calibration over tensors (ReCat) recovers missing data by tensor interpolation over the coil and acquisition dimensions. Coil compression is achieved using a new method based on multilinear singular value decomposition (MLCC). ReCat is compared with iterative self-consistent parallel imaging (SPIRiT) and profile encoding (PE-SSFP) reconstructions. Results: Compared to parallel imaging or profile-encoding methods, ReCat attains sensitive depiction of high-spatial-frequency information even at higher R. In the brain, ReCat improves peak SNR (PSNR) by 1.1$\pm$1.0 dB over SPIRiT and by 0.9$\pm$0.3 dB over PE-SSFP (mean$\pm$std across subjects; average for N=2-8, R=8-16). Furthermore, reconstructions based on MLCC achieve 0.8$\pm$0.6 dB higher PSNR compared to those based on geometric coil compression (GCC) (average for N=2-8, R=4-16). Conclusion: ReCat is a promising acceleration framework for banding-artifact-free bSSFP imaging with high image quality; and MLCC offers improved computational efficiency for tensor-based reconstructions.

physics.med-ph↗

arXiv2017

Generalized Approximate Message-Passing Decoder for Universal Sparse Superposition Codes

Sparse superposition (SS) codes were originally proposed as a capacity-achieving communication scheme over the additive white Gaussian noise channel (AWGNC) [1]. Very recently, it was discovered that these codes are universal, in the sense that they achieve capacity over any memoryless channel under generalized approximate message-passing (GAMP) decoding [2], although this decoder has never been stated for SS codes. In this contribution we introduce the GAMP decoder for SS codes, we confirm empirically the universality of this communication scheme through its study on various channels and we provide the main analysis tools: state evolution and potential. We also compare the performance of GAMP with the Bayes-optimal MMSE decoder. We empirically illustrate that despite the presence of a phase transition preventing GAMP to reach the optimal performance, spatial coupling allows to boost the performance that eventually tends to capacity in a proper limit. We also prove that, in contrast with the AWGNC case, SS codes for binary input channels have a vanishing error floor in the limit of large codewords. Moreover, the performance of Hadamard-based encoders is assessed for practical implementations.

cs.IT↗