arXiv subjects

Taekyung Kim

Publications and source records attributed to Taekyung Kim.

At least 19 recordsLinked to original sources

arXiv2026

Omni-Modal LLMs Know More Than They Say

Omni-modal large language models (Omni-LLMs) extend LLMs to process heterogeneous modalities (e.g., vision, audio, and IMU signals) alongside language, enabling a broader range of multimodal tasks. Yet we find that these models often fail to express information from non-textual modalities even when they perceive it correctly. We term this disconnect knows-but-cannot-say. To investigate this phenomenon, we introduce Token-level Modality Alignment (TMA), which measures the alignment between individual modality representations and response representations. Our analysis shows that non-textual representations are less aligned with response tokens and receive less attention than textual representations, suggesting a potential explanation for uneven modality preferences. Motivated by these findings, we propose OmniCatalyst, a simple perceive $\rightarrow$ infer strategy that complements the original modality inputs with self-generated text, making non-textual knowledge more accessible. Across seven Omni-LLMs, OmniCatalyst consistently improves performance (+7.29%) and yields more balanced modality preferences, with further gains from diverse training strategies. Further analysis demonstrates that the knowledge extraction phases increase TMA, suggesting that improved alignment makes non-textual information more accessible during response generation. In summary, our findings uncover how Omni-LLMs can better express modality knowledge through language. Our code is available at https://github.com/naver-ai/OmniLLM.

cs.CL↗

arXiv2026

Learning Transferable World-Action Models from Task-Paired Human Videos

World-action models inherit broad visual and physical priors from video pre-training. However, standard within-video next-frame prediction allows them to rely on appearance shared between the context and future, rather than learning interactions that transfer across scenes and embodiments. We introduce WATCH, a cross-video pre-training that learns from task-paired human demonstrations collected across diverse environments. Conditioned on the video and actions of one demonstration, the model jointly predicts another demonstration's actions and future frames from its current observation. Predicting one demonstration from another encourages the model to learn task-relevant interactions that transfer across scenes and demonstrators. Our pre-training shows that these capabilities can be learned from readily scalable human-human pairs alone, without requiring matched human-robot demonstrations. Analyses before robot post-training show improved motion alignment across embodiments and better action and video prediction on unseen robots, with representations closer to those learned from robot pairs. After robot post-training, WATCH improves MimicDroid in-context success from 40.9% to 68.3% with Cosmos Policy and from 59.8% to 77.7% with Cosmos3, with the largest gains on the unseen levels. Gains persist under context-free post-training across robot embodiments. WATCH outperforms within-video pre-training and co-training baselines, supporting cross-video prediction as a scalable approach to learning transferable world-action representations.

cs.CV↗

arXiv2026

Sparse Feature Policy Unlearning Mitigates State Hallucination in Vision-Language-Action Models

Vision-Language-Action (VLA) models have shown strong generalization in robotic manipulation by leveraging rich representations from pretrained vision-language models. However, their deployment in real-world environments remains limited by recurring unreliable behaviors. In this work, we study state hallucination, a recurring failure pattern in which a VLA continues acting as if an unrealized robot-object state had been achieved. Our analyses find that state hallucination coincides with weakened attention to task-relevant visual regions, and a mechanistic interpretation via sparse autoencoders reveals that hallucination-associated sparse features are activated when these failures occur. Based on this analysis, we propose SOUL (Sparse feature pOlicy UnLearning), which selectively unlearns policy knowledge associated with state hallucination behaviors, where sparse features identified from hallucination failures and successful behaviors serve as explicit forgetting and retention targets, respectively. Experiments across VLA architectures in simulated and real-world environments show that our method substantially reduces hallucinated failures and improves task success without substantially compromising the existing manipulation capabilities. These results suggest that interpretable feature analysis provides a practical basis for selectively modifying undesirable knowledge in robot policies.

cs.RO↗

arXiv2026

DODGER: Safety-Guided Reinforcement Learning for Robot Navigation Among Dynamic Obstacles

Robots operating in human-centered environments must safely navigate among multiple dynamic obstacles to avoid collisions with people and surrounding infrastructure. Control barrier functions (CBFs) provide an effective mechanism for safety filtering, and recent CBF-based reinforcement learning (RL) methods embed such safety information into learned policies. However, executing only safety-filtered actions during training can restrict policy exploration, a limitation that becomes particularly consequential in dynamic scenes where safety depends on relative robot-obstacle motion. We propose DODGER, a safety-guided RL framework that directly executes policy-generated actions to drive training rollouts while using CBF-filtered references and constraint violations to shape the policy toward collision-avoidance behavior. We evaluate DODGER through a Dubins-car safety analysis and demonstrate goal-directed navigation among multiple dynamic obstacles in full-order humanoid simulation and real-world humanoid experiments using LiDAR-based perception, without a runtime safety filter.

cs.RO↗

arXiv2026

BR-MPPI: Barrier-Rate Guided MPPI for Enforcing Multiple Inequality Constraints with Learned Signed Distance Fields

Model Predictive Path Integral (MPPI) control provides a sampling-based framework for optimal control, while Control Barrier Functions (CBFs) provide a principled means of enforcing safety constraints. We introduce BR-MPPI, which integrates CBF-like conditions into MPPI's control sampling procedure. CBFs impose inequality constraints that bound the rate of change of barrier functions using a class-K function of the barrier value. We instead impose the CBF condition as an equality constraint using a parametric linear class-K function and augment the system state with its parameter. The parameter's time derivative serves as an additional control input optimized by MPPI. We further design a cost function that promotes parameter values consistent with Nagumo's condition at the safe-set boundary, thereby encouraging safety. The resulting multiple state- and control-dependent equality constraints pose a challenge for random control sampling. We address this through state transformations and control projections inspired by manifold path planning that map sampled controls onto the constraint manifold. We also incorporate learned signed distance fields to represent robot geometry and reduce computation time. Simulations demonstrate improved sample efficiency over vanilla MPPI and higher navigation success rates across five robot models compared with safety-oriented MPPI variants. Hardware experiments on a quadrotor further demonstrate the method's ability to navigate constrained environments near safe-set boundaries.

cs.RO↗

arXiv2026

Predictive Semantic Safety: From Visual Physical Reasoning to Safety-Critical Control

Physical interactions can create future hazards that are not apparent from the robot's current geometric surroundings. We present a framework termed Predictive Semantic Safety (PSS), which connects visual physical reasoning to backup-based safety filtering. A vision-language model (VLM) predicts physical events and their timing or directly predicts object displacements. An explicit motion model converts event hypotheses into object trajectories. Split conformal prediction calibrates position errors jointly across specified objects, observation times, and future times; geometric shape bounds convert the resulting position regions into predicted object occupancy. PSS evaluates a prescribed backup maneuver against this occupancy and derives input-affine constraints for minimally modifying the nominal input while preserving backup feasibility under the robot dynamics and input limits. MuJoCo experiments with a Unitree Go1 consider falling fixtures, impact-driven support loss, and contact propagation. PSS achieves a safe episode rate of 99.3%, compared with 43.3% for a Backup Control Barrier Function baseline that only uses current obstacle geometry.

cs.RO↗

arXiv2026

DiffCVaR: Reinforcement Learning for Risk Adaptation via Differentiable CVaR Barrier Functions

Planning through crowded environments under uncertain obstacle motions remains difficult, as stochastic interactions often induce overly conservative behavior or reduced efficiency. To address this challenge, we propose an end-to-end risk adaptation framework for crowd navigation under obstacle-motion uncertainty modeled by a Gaussian mixture model. The framework combines reinforcement learning~(RL) with a differentiable quadratic-program safety layer based on Conditional Value-at-Risk~(CVaR) barrier functions, jointly learning nominal control input, risk level, and safety margin and enforcing explicit probabilistic safety constraints. This design enables context-aware adaptation, promoting efficient behavior while invoking caution only when necessary. We conduct extensive evaluations in dynamic, uncertain, and crowded environments across varying obstacle densities and robot models, and further assess generalization under three out-of-distribution cases. Comparisons across optimization-based, RL-based, and integrated RL and optimization methods are provided, and the proposed method is shown to deliver the strongest overall performance in safety, efficiency, and generalization under uncertainty.

cs.RO↗

arXiv2026

Backup-Based Safety Filters: A Comparative Review of Backup CBF, Model Predictive Shielding, and gatekeeper

This paper revisits three backup-based safety filters -- Backup Control Barrier Functions (Backup CBF), Model Predictive Shielding (MPS), and gatekeeper -- through a unified comparative framework. Using a common safety-filter abstraction and shared notation, we make explicit both their common backup-policy structure and their key algorithmic differences. We compare the three methods through their filter-inactive sets, i.e., the states where the nominal policy is left unchanged. In particular, we show that MPS is a special case of gatekeeper, and we further relate gatekeeper to the interior of the Backup CBF inactive set within the implicit safe set. This unified view also highlights a key source of conservatism in backup-based safety filters: safety is often evaluated through the feasibility of a backup maneuver, rather than through the nominal policy's continued safe execution. The paper is intended as a compact tutorial and review that clarifies the theoretical connections and differences among these methods.

cs.RO↗

arXiv2026

Distributed Model-Based Diffusion For Scalable Multi-Robot Trajectory Optimization

Trajectory optimization for multi-robot systems remains a critical challenge, particularly when navigating highly non-convex, non-linear, and non-differentiable environments. While Model-Based Diffusion (MBD) has recently emerged as a promising sampling-based optimization paradigm for single-robot trajectory generation, extending it to multi-robot systems results in a centralized, high-dimensional inference problem that (i) suffers from poor sample efficiency due to the curse of dimensionality and (ii) requires global access to all robots' dynamics, constraints, and objectives. To address this, we propose Distributed Model-Based Diffusion (DMBD), a distributed server-robot method that decomposes the reverse diffusion process into local conditional reverse diffusion processes. This decomposition enables each robot to iteratively perform denoising independently within its own control subspace while conditioning on the current trajectory estimates of the other robots that are aggregated and broadcast by the server. Extensive simulations in goal swapping, multi-floor coverage, parking, and rush-hour scenarios demonstrate that DMBD achieves strong scalability, solving many challenging coordination tasks with sub-second computation time and outperforming existing baselines.

cs.RO↗

arXiv2026

Isotropic Embedding Perturbations for Robust Vision Language Encoders

Data augmentation is fundamental to training modern deep vision and multimodal models. While individual methods, such as RandAug, CutMix, Mixup, RandErase, and DropPath, offer strong regularization effects, their combined use has saturated in performance due to overlapping functionalities, and aggressive pixel-level manipulations may disrupt delicate cross-modal alignment. This saturation motivates the search for a new augmentation axis within the embedding space rather than the input space. We introduce Aether, a simple plug-in method that applies diffusion-style random perturbations in the embedding space via controlled alpha-mixing, specifically designed to provide isotropic regularization that remains semantically consistent. Inspired by feature-space perturbations in language models and image degradation in generative pretraining, Aether induces mild yet effective perturbations that smooth the representations without compromising the fine-grained structural information required for strong vision-language encoders. Across diverse architectures and across multiple recognition tasks, Aether delivers consistent gains over the advanced recipe combining CutMix, Mixup, DropPath, and RandAug---a level of improvement rarely observed with modern augmentation alternatives. Notably, Aether demonstrates superior effectiveness in multi-modal alignment, succeeding where traditional pixel-space augmentations fail by providing a stable, isotropic regularization signal that respects the integrity of the high-dimensional feature space.

cs.CV↗

arXiv2026

OcclusionCBF: Backup Control Barrier Functions for Safe Navigation Among Hidden Dynamic Obstacles

Robots navigating under occlusion may enter states from which no admissible input can avoid a dynamic obstacle once it becomes visible. We present OcclusionCBF, a safety filter that extends backup control barrier functions to reachable-occupancy predictions for potentially hidden dynamic obstacles in occluded regions. The method certifies a prescribed backup rollout against collision-inflated occupancy and a verified terminal set, yielding affine constraints for minimally invasive quadratic-program filtering. We establish recursive feasibility of the resulting safety filter, and collision avoidance for every hidden-obstacle motion covered by the occupancy prediction. Randomized benchmarks, MetaUrban simulations, and hardware experiments demonstrate improved task success over reactive and occlusion-aware predictive baselines with millisecond-scale computation.

cs.RO↗

arXiv2026

Beneath the Surface of Chains-of-Thought: A Mechanistic Interpretation of Reasoning Operations in LLMs

Reasoning in large language models unfolds through diverse functional operations, such as problem formulation, goal decomposition, and deduction. Although these operations are explicitly distinguished in text, little is known about how they are geometrically organized in representation spaces. To this end, we investigate whether distinct reasoning operations exhibit corresponding geometric structure in hidden representations. We find that operations are separable in held-out representations, with separability peaking in middle layers, and verify that this structure is not explained by lexical or positional confounds. Across layers, token-wise operation-alignment becomes more distributed over spans, while identical surface tokens are represented differently depending on the operation of its surrounding chunk. Attention-masking interventions further show that operation-aligned representations at chunk onset depend on preceding reasoning context. Consequently, our work demonstrates that language models maintain representational correspondence between linguistic reasoning expressions and their internal geometric structures. Code and project materials are available at https://github.com/naver-ai/beneath-cot.

cs.CL↗

arXiv2026

Safety-aware Model Predictive Path Integral Control with Signal Temporal Logic

Safety-aware motion planning remains a challenge in robotics, especially when missions are time-critical and are under complex specifications. In this paper, we propose safety-aware-stl-mppi, a computationally efficient sampling-based receding-horizon planning framework designed to promote satisfaction of constraints expressed in Signal Temporal Logic (STL). Our approach encodes discrete-time STL formulas into candidate time-varying control barrier functions (CBF), which are integrated into a model predictive path integral (MPPI) controller. Our method inherits the benefits of low computational cost from an efficiently parallelizable sampling based planner and utilizes CBF for constraints expressed in STL. We compare against several MPPI baselines using four artificial Mars Rover planning case studies with a diverse environment and cost setups, where we show our method consistently achieving high safety and efficiency. We show a quadcopter planning experiment with NVIDIA Isaac Lab.

cs.RO↗

arXiv2026

Distributionally Robust Safety Under Arbitrary Uncertainties: A Safety Filtering Approach

We study how to ensure probabilistic safety for nonlinear systems under distributional ambiguity. Our approach builds on a backup-based safety filtering framework that switches between a high-performance nominal policy and a certified backup policy to ensure safety. To handle arbitrary uncertainties from ambiguous distributions, i.e., where the distribution is not of specific structure and the true distribution is unknown, we adopt a distributionally robust (DR) formulation using Wasserstein ambiguity sets. Rather than solving a high-dimensional DR trajectory optimization problem online, we exploit the structure of backup-based safety filtering to reduce safety certification to a one-dimensional search over the switching time between nominal and backup policies. We then develop a sampling-based certification procedure with finite-sample guarantees, where empirical failure probabilities are compared against a Wasserstein-inflated threshold. We validate our method across three systems, from a Dubins vehicle to a high-speed racing car and a fighter jet, demonstrating the broad applicability and computational efficiency.

cs.RO↗

arXiv2026

Is Your Safe Controller Actually Safe? A Critical Review of CBF Tautologies and Hidden Assumptions

This tutorial provides a critical review of the practical application of Control Barrier Functions (CBFs) in robotic safety. While the theoretical foundations of CBFs are well-established, I identify a recurring gap between the mathematical assumption of a safe controller's existence and its constructive realization in systems with input constraints. I highlight the distinction between candidate and valid CBFs by analyzing the interplay of system dynamics, actuation limits, and class-K functions. I further show that some purported demonstrations of safe robot policies or controllers are limited to passively safe systems, such as single integrators or kinematic manipulators, where safety is already inherited from the underlying physics and even naive geometric hard constraints suffice to prevent collisions. By revisiting simple low-dimensional examples, I show when CBF formulations provide valid safety guarantees and when they fail due to common misuses. I then provide practical guidelines for constructing realizable safety arguments for systems without such passive safety. A crowd-navigation simulation study further illustrates that CBF-derived reward shaping in reinforcement learning can improve empirical behavior without establishing formal safety. The goal of this tutorial is to bridge the gap between theoretical guarantees and actual implementation, supported by an open-source interactive web demonstration that visualizes these concepts intuitively.

cs.RO↗

arXiv2026

SEAMLiS: Visibility-Aware Safety for Perception-Limited Multi-Robot Exploration

Autonomous exploration in unknown environments is typically driven by informative frontiers, viewpoints, or trajectories, while local safety controllers avoid obstacles represented in the current map. Under finite sensing range and limited field of view, this separation can be unsafe: an exploration stack may plan optimistically through unobserved space and steer the sensor toward information gain rather than along the direction of motion, causing hidden obstacles to be detected too late for bounded-actuation avoidance. This paper presents SEAMLiS (Safe Exploration for Autonomous Multi-Robot Systems Under Limited Sensing), a modular execution-layer safety framework for decentralized multi-robot exploration. SEAMLiS preserves the upstream exploration stack, including the goal allocator and local planner, and enforces safety at the execution layer through perception-aware attitude and positional filters. A gatekeeper-based attitude filter switches between a visibility-promoting yaw policy and a velocity-tracking backup policy to preserve visibility of the critical known-free/unknown boundary with sufficient braking margin. A Control Barrier Function (CBF)-based positional filter then avoids known obstacles, newly detected obstacles, and other robots. We provide sufficient collision-avoidance conditions and validate the framework in randomized simulation, Isaac Sim, and Crazyflie hardware experiments. Results show collision-free exploration across tested single- and multi-robot settings while retaining much of the efficiency of visibility-promoting yaw control.

cs.RO↗

arXiv2026

Retrieve, Don't Retrain: Extending Vision Language Action Models to New Tasks at Test Time

Extending a vision-language-action (VLA) policy to a new task typically requires task-specific teleoperated demonstrations and per-task fine-tuning, making adaptation costly in both data collection and compute. In this paper, we show that this target-side per-task adaptation cost can be replaced by retrieval. Our retrieval-augmented policy is trained once on paired demonstrations from the target embodiment (query) and a cheaper embodiment (pool, e.g., human-hand video), then frozen. New tasks are added at deployment by appending pool-side demonstrations to a retrieval pool. The frozen policy conditions on retrieved trajectories at every control step, so new tasks are absorbed by indexing data rather than updating parameters. Fine-tuning is needed only to take on a new, unseen embodiment, not for each new task. We show that retrieval improves policies beyond a specific backbone, including standard VLA policies, but its effect is especially pronounced in Cosmos Policy, a video-generation-based world-action model (WAM). In this setting, retrieval supplies coarse task progression, while the WAM's future-image objective provides an additional visual consistency signal that strengthens the retrieval-conditioned actions. On PushT, we study how retrieval provides a reusable high-level motion prior for cross-embodiment generalization to unseen goal angles, while on RoboTwin 2.0 our method outperforms cross-embodiment baselines on unseen tasks, and we additionally demonstrate the method on a real robot.

cs.RO↗

arXiv2026

Learning to Adapt Control Barrier Functions Under Epistemic and Aleatoric Uncertainty

Control barrier functions (CBFs) provide a tractable mechanism for enforcing safety constraints in robotic systems, but their practical performance depends strongly on the choice of class-K function parameters. Under input constraints, conservative parameters often preserve feasibility at the cost of slow progress, whereas aggressive parameters can make the CBF-based optimization infeasible or unsafe. This paper proposes Online Adaptive CBF (OA-CBF), a framework for adapting CBF parameters at runtime. We introduce the notion of locally validated CBF parameters, which certify candidate parameters over a finite prediction horizon, and show that safety is preserved when such validation is maintained over successive update intervals. To identify locally validated parameters efficiently, OA-CBF trains a probabilistic ensemble neural network to evaluate queried CBF parameters rather than directly predict a single parameter. A graph-attention encoder represents variable-size obstacle environments, an epistemic uncertainty gate calibrated by conformal prediction rejects unreliable predictions, and a distributionally robust CVaR condition screens aleatoric risk. Among the verified candidates, OA-CBF selects the parameter with the best predicted progress metric and applies it through either an MPC-CBF or CBF-QP safety filter. Simulation studies on dynamic unicycle, planar and three-dimensional quadrotor, kinematic bicycle, and VTOL quadplane benchmarks show that OA-CBF reduces the conservatism of fixed-parameter CBF controllers while maintaining low collision and infeasibility rates.

cs.RO↗