Action
-
From Reasoning to Action: Benchmarking Frontier AI Models on LIBERO-10 Robot Manipulation Tasks
2026-10-21 17:11:52May Espinola, Qifan Zhang, Yu XiangAbstract
Recent advances in frontier AI models have demonstrated remarkable capabilities in visual understanding, reasoning, planning, and tool use, raising the possibility of using general-purpose AI models for robot manipulation without task-specific training. However, their manipulation capabilities have not been systematically evaluated on standardized robotics benchmarks. In this report, we present a unified closed-loop manipulation framework for evaluating frontier AI models on the LIBERO-10 benchmark. Our framework enables models to control a robot directly through visual observations, robot proprioception, and structured tool calls, primarily by specifying target end-effector poses and gripper commands, without task-specific fine-tuning or pretrained manipulation skills. We evaluate six frontier AI models, including GPT-6 Astra, GPT-6.1 Sol, GPT-6 Luna, Claude Opus 5, Claude Opus 5.5, and Gemini-4 Argon, across three reasoning effort levels, resulting in 18 configurations and 1,800 evaluation episodes. Our results show that GPT-6 Astra achieves the highest success rate of 44%, followed by Gemini-4 Argon at 43%. Increasing reasoning effort generally improves manipulation performance, but the benefits vary across models and come at additional inference cost. We also observe substantial differences across tasks, with models performing relatively well on object placement tasks but struggling with precise spatial arrangements and complex multi-step interactions. These findings demonstrate both the potential and limitations of frontier AI models for robot manipulation, highlighting the challenges of translating reasoning capabilities into reliable physical actions. We release our evaluation framework and experimental results to support reproducible benchmarking and future research on reasoning-based robot manipulation.
Abstract (translated)
URL
https://arxiv.org/abs/2610.30051
PDF
-
Does Centralized Learning Help Embodied Agents Coordinate, and When Can a Tabular Study Tell?
2026-10-21 17:09:52Muhammad Ahmed Atif, Nehal Naeem Haji, Muhammad Ebad Atif, Mohammad Shahid ShaikhAbstract
Tabular studies are a common way to compare independent and centralized multi-agent learners, because they avoid function approximation. A natural expectation is that centralized learning, which learns joint actions, helps agents coordinate, especially when their bodies make solo success hard. Earlier work tests this only in parts: it measures coordination for one learning method, compares methods with the agents' bodies fixed, or varies speed but reports only how well the team did. We test it in a predator-prey environment, where teammates cooperatively hunt together while two teams compete. We compare independent Q-learning (IQL) with central Q-learning (CQL) in all four predator and prey pairings, under limits on speed and stamina, on a 5x5 and an 8x8 grid trained for 200,000 episodes each, and check that the agents learn on both. We measure coordination directly, with predictions for the main measurement fixed in advance. In this task, where one predator can capture prey alone, we find three things. First, CQL costs its agents in both roles on both grids, and its predators coordinate no more than IQL's. Second, its one benefit comes from acting on the best joint action: this cut capture time by 42% on the small grid with faster prey and unlimited stamina, but helped nowhere on the larger grid, where the best joint action had mostly never been tried. Third, on the larger grid a tie-breaking default decided which method won, even though the agents had passed a learning check. A tabular comparison can therefore be trusted only if it also checks how much of the state space the agents covered and implements the methods alike. We propose simple checks for both.
Abstract (translated)
URL
https://arxiv.org/abs/2601.17454
PDF
-
SWE-DIVE: Multi-Turn Coding Agent Evaluation with Dynamic User Simulation
Abstract
Coding agents are typically evaluated using one-shot, well-formulated prompts. Benchmarks such as SWE-bench use existing GitHub issues as input for a coding agent and evaluate whether the agent successfully solves the task. However, in real-world use cases, prompts are often not as well-formulated or structured as existing GitHub issues. Additionally, a coding agent session is not limited to a single prompt; it typically involves a multi-turn interaction in which the user iteratively adds new tasks and requirements that the agent must fulfill. To benchmark these abilities, we developed SWE-DIVE, a dynamic, interactive validation and evaluation framework for benchmarking coding agents in a multi-turn environment. To achieve this, we introduce a user simulator that can simulate different behavioral patterns and communicate with the coding agent. We also introduce a conversation validator that evaluates the user simulator's outputs to determine whether they fulfill the initial requirements and guide the agent toward successfully solving the task. The entire framework is embedded in a Harbor environment, which orchestrates the coding agent, user simulator, and conversation validator, and executes the test environment to determine whether the agent successfully solves the task. Our results and source code are available in our GitHub repository (this https URL).
Abstract (translated)
URL
https://arxiv.org/abs/2610.30047
PDF
-
SP-HLM: Sparse-Pose-Driven Motion Generation and Whole-Body Control for Humanoid Loco-Manipulation
2026-10-21 17:03:34Zhaohui Yang, Quanbao Lin, Chengqi Shi, Zhenglang Weng, Jicheng Lin, Boyu Chen, Tengjie Zhu, Guanzhu Ren, Jiahao Xu, Yang Feng, Xiaokang Yang, Yao MuAbstract
Humanoid loco-manipulation requires an effective interface between high-level task policies and whole-body control. Existing interfaces typically rely on a single action representation for both task intent and whole-body coordination. This may burden the high-level policy or leave one aspect underspecified. We present SP-HLM, a framework built on a sparse-pose-driven interface. The interface combines five task-critical body poses with a learned latent motion prior to represent spatial intent and whole-body coordination in complementary forms. SP-VLA, a vision-language-action model, first predicts task-relevant sparse poses and finger joint positions from multimodal observations and language instructions using asynchronous action-chunk inference. Given these sparse targets, SP-MGM, a motion generation model trained through two-stage distillation, reconstructs coordinated full-body motion in real time. A pretrained motion encoder maps this motion to a compact latent representation. SP-GMT, a general motion tracker, then integrates the explicit sparse targets with this implicit motion prior for whole-body control. Experiments in simulation and on a real humanoid robot demonstrate improved motion tracking, whole-body coordination, and loco-manipulation success over existing approaches. Additional results and videos are available on our \href{this https URL}{\textcolor{blue}{project page}}.
Abstract (translated)
URL
https://arxiv.org/abs/2610.30044
PDF
-
TARSphere: A Robot-Oriented Full-Stack Framework for Long-Horizon Embodied Agents in Evolving Physical and Social Environments
2026-10-21 17:02:52Yicheng Xu, Zhikai Xu, Ziyuan Chen, Yuhang Dong, Tao Zhang, Zhucun Xue, Yong Liu, Jiangning ZhangAbstract
Robots must act under physical and social constraints while adapting to changes in their environment and human needs. Existing embodied-agent systems often focus on specified goals or abstract execution into symbolic actions, limiting evaluation in an evolving world. We introduce TARSphere, a robot-oriented full-stack framework for physically and socially grounded long-horizon embodied agents. TARSphere-Infra runs robot policies in a physics simulator with embodied NPCs and constrained dialogue. An event-driven finite-state machine (FSM) uses completed actions and delivered information to update task states and evaluate outcomes. TARSphere-Harness, its embodied brain harness, lets a frozen multimodal large language model (MLLM) act through multi-level executable skills, using observations, hierarchical memory, and execution feedback to revise its decisions. On this foundation, agent-in-the-loop curation aligns task semantics, causal dependencies, and scene geometry to construct TARSphere-Bench: 360 episodes across 30 indoor scenes. Five paired conditions assess perceptual robustness, disruption adaptation, task update, human coordination, and intent understanding for navigate-and-report and fetch-and-deliver tasks. Our evaluation of 8 MLLMs shows that general-purpose models can outperform embodied models, while task updates and human interaction remain difficult. Frequent collisions and falls further motivate evaluating execution safety alongside task success, even with a shared action interface and decision budget.
Abstract (translated)
URL
https://arxiv.org/abs/2610.30043
PDF
-
Agent Behavior Maps: Locating Agents by What They Would Do, Without Running Them
2026-10-21 16:44:02Shota Fujikawa, Takuya Tamura, Kosuke Akimoto, Daichi Haraguchi, Masafumi OyamadaAbstract
LLM agents with similar success rates can follow different action sequences. An Agent Behavior Map represents LLM-harness combinations using centered action log-likelihoods on shared reference trajectories, without requiring new environment rollouts from the target agent. We derive error bounds relating squared coordinate distances to trajectory KL under explicit assumptions, and across eight cross-scored models the distances track the directly measured trajectory KL in rank order. Across 100 LLMs and four environments, ALFWorld-based coordinates support performance prediction and identify behavioral similarities. In ALFWorld, these similarities persist after controlling for success rates and model size, including across model families. As an extension, harness-component displacements approximately compose, with accuracy varying across components and environments.
Abstract (translated)
URL
https://arxiv.org/abs/2610.30014
PDF
-
DONIC: Generalist Dexterous Bimanual Tracking Control from Human Demonstrations
2026-10-21 16:36:02Mingze Sun, Ziang Cao, Yuan Fei, Qiankuan Tao, Zihan Wang, Tianqi Liu, Junhao Chen, Haiwen Diao, Fangzhou Hong, Zhaoxi Chen, Ziwei LiuAbstract
Despite recent progress in dexterous reference tracking, existing models focus primarily on single-hand manipulation and a limited range of interactions, leaving the bimanual dexterous behaviors common in real life unaddressed. The scarcity of contact-rich bimanual manipulation data and the increased complexity of coordinating contacts make developing a generalist bimanual tracker particularly challenging. Therefore, we present 1) DONIC, the first generalist bimanual tracker jointly trained across operation categories and human-motion sources, together with 2) BiHOI, a multi-view dataset of 1,575 sequences spanning 229 objects and emphasizing sustained bimanual interaction, where reconstructed interactions are retargeted into robot references. DONIC combines proximal policy optimization with behavioral cloning in a feedback loop that recycles verified successful executions into training demonstrations. An object-centered per-hand reward maintains object-motion feedback while supervising each hand according to its reference contact role. Clip-Centered Advantages remove reference-dependent advantage offsets while preserving within-trajectory differences. Together, these designs enable one single policy to learn coordinated manipulation across heterogeneous references. Without finetuning, a single checkpoint achieves 86.79% success on unseen trajectories of training objects and 57.32% on held-out objects, surpassing the state-of-the-art bimanual tracking baselines by 77.36 and 45.73 percentage points.
Abstract (translated)
URL
https://arxiv.org/abs/2610.29997
PDF
-
Remembering Through Change: Provenance-Preserving Hypergraph Memory for Multimodal Agents
2026-10-21 16:34:28Jielong Tang, Ling-An Zeng, Xujie Yuan, Xiao Dong, Zhiqian Xia, Qingfeng Li, Shixiong Kai, Kaichao Liang, Rong Yao, Jian YinAbstract
Long-term memory for multimodal agents remains challenging, as it must preserve fine-grained visual evidence while tracking entity evolution across sessions. Existing abstraction-based memories compress images into text, often discarding visually discriminative details, whereas pixel-preserving memories retain raw observations but struggle to reliably associate diverse observations with persistent entity identities and track their evolving states. To address these limitations, we propose MHG-Mem, a multimodal hypergraph memory that jointly organizes entity identity, state evolution, and provenance to visual evidence. Specifically, MHG-Mem represents distinct entity mentions and their observed facts as individual nodes, grouping identity-consistent mentions and facts through occurrence hyperedges. It then synthesizes aspect-specific evolution states from these grouped observations and links them to supporting entities, facts, and source evidence through evolutionary and evidential hyperedges. To navigate this structured memory, we introduce hypergraph-based agentic search, which adaptively selects an entry layer, traverses typed hyperedges across memory levels, and inspects source observations when fine-grained verification is required. These designs support both detailed visual verification and reasoning over cross-session state changes. Extensive experiments across three benchmarks demonstrate the effectiveness of MHG-Mem, improving performance by 8.20-9.98% over the strongest baseline.
Abstract (translated)
URL
https://arxiv.org/abs/2610.29995
PDF
-
Towards English - Mizo Legal Machine Translation: Corpus Creation, Analysis, and Benchmarks
2026-10-21 16:11:01Telem Joyson Singh, Sanabam Boyshana, Sanasam Ranbir Singh, Priyankoo SarmahAbstract
Indian court proceedings are held in English, so speakers of languages with limited translation technology have limited access to them. We present the first English Mizo parallel dataset of Indian court judgments. It contains 13,168 sentence pairs from the Gauhati High Court and the Supreme Court of India. Professional native Mizo translators translated complete documents; sentence pairs were manually aligned, and two independent reviewers checked the results. Our study finds that English words are borrowed in a very selective way: official and legal terms are almost always kept in English, but words for everyday things like actions, situations, and results are usually expressed in native Mizo. Mizo diacritics are also used inconsistently across the corpus. When we fine tune the NLLB 200 1.3B model on our training data, it scores 27.47 BLEU on a balanced legal test set of 1,000 sentence pairs, an improvement of 7.16 BLEU over the base model. However, it still uses English words about three times more often than a human translator does.
Abstract (translated)
URL
https://arxiv.org/abs/2610.29943
PDF
-
DewHand: Direct Metric World-Space Hand Motion Reconstruction from Egocentric Videos
2026-10-21 16:08:38Naru Suzuki, Satoshi Ikehata, Takehiko Ohkawa, Yoichi Sato, Shuhei KuritaAbstract
Recovering metric world-space hand motion from egocentric video supports understanding human interactions and transferring them to embodied agents. Three challenges complicate this task: multi-stage pipelines propagate errors from independently estimated hand and camera motion; monocular observations leave physical scale ambiguous; and occlusion and out-of-view hands admit multiple plausible motion trajectories. We present DewHand, a unified model for direct joint reconstruction of world-space hand and camera motion. To avoid compounding errors from multi-stage estimation, DewHand jointly predicts hand and camera motion in a shared world coordinate system within a single-stage pipeline. To resolve metric-scale ambiguity, it combines scene-level geometric features from VGGT-$\Omega$ with metric-space supervision, while hand-specific features preserve fine-grained articulation. To handle occlusion and out-of-view hands, we formulate reconstruction as conditional motion generation with flow matching. Under the same architecture, this generative formulation outperforms direct regression in both accuracy and temporal smoothness with only two sampling steps. Experiments on HOT3D and ARCTIC demonstrate improved world-space accuracy and temporal consistency over state-of-the-art methods. DewHand reduces rigid-aligned world-space error relative to StableHand from 32.9 to 28.4 mm on HOT3D and from 76.2 to 36.2 mm on ARCTIC. It also recovers camera scale within 10% on two thirds of HOT3D clips, versus about a quarter for SLAM. Moreover, in zero-shot evaluations on HoloAssist, H2O, and TACO, captured with different devices, DewHand achieves the lowest rigid-aligned world-space error (WA-MPJPE) on each dataset.
Abstract (translated)
URL
https://arxiv.org/abs/2610.29936
PDF
-
ToM2Vec: Continuous Theory-of-Mind Tokens for Multi-Agent Coordination
2026-10-21 16:05:54Nithin Sivakumaran, Ali Payani, Hyunji Lee, Mohit Bansal, Elias Stengel-EskinAbstract
Forming an accurate theory-of-mind (ToM) is key to successful multi-agent collaboration and requires reasoning under uncertainty to capture multiple possible partner states. This is hard to do in the text representation space native to large language model (LLM) agents, which more easily describe a single partner state. To address this challenge, we introduce ToM2Vec, which learns to form abstract theories-of-mind in a rich latent space via a fixed budget of continuous reasoning tokens. During training, supervision for these tokens comes from the partner's actual thoughts, which the ToM tokens are trained to abstract and match. At test-time, ToM tokens are generated by each agent and act as conditioners for that agent's thoughts and actions. Across several environments covering collaborative, competitive, and mixed-motive settings (Overcooked-AI, multiparty negotiation, Decrypto, Hanabi, and Collab Escape/Capture), we evaluate ToM2Vec's ability to support multi-agent interactions and collaboration, finding consistent improvements over the base model and a variety of competitive trained and prompted baselines. We show that latent reasoning is critical, with performance dropping when reasoning is expressed in text. Moreover, we show that ToM2Vec training generalizes to partner model families not seen at train-time, including a proprietary model (GPT-5.4). In our analysis, we provide evidence that the latent tokens encode a ToM, rather than task-specific information: ToM2Vec's tokens improve the agent's ability to directly answer questions about partner beliefs, desires, and next actions.
Abstract (translated)
URL
https://arxiv.org/abs/2610.29934
PDF
-
Your Benchmark Is an Empirical Measure Over Difficulty
2026-10-21 16:02:37Yifan Sun, Naicheng Yu, Jiawen Gong, Jingyan Shen, Huan ZhangAbstract
A benchmark is not merely a collection of test questions; it naturally induces an empirical measure over difficulty. In this work, we develop this measure-based view of benchmark evaluation. This perspective allows us to identify and formalize two central failure modes: measure mismatch, where the benchmark difficulty distribution differs from the intended difficulty distribution, and coverage gaps, where uncovered regions along the difficulty axis prevent benchmarks from detecting meaningful localized training progress. We propose two diagnostic metrics that capture these two failure modes and provide rigorous theoretical guarantees. Empirically, we analyze 25 popular benchmarks spanning 7 domains using carefully estimated item difficulties. Our results show that many benchmarks implicitly overweight certain difficulty regions, so their leaderboard conclusions should be interpreted with caution; others leave substantial portions of the difficulty axis uncovered, limiting their ability to monitor the model improvement process. Finally, we turn diagnosis into action by introducing BenchFill, a simple difficulty-targeted item rewriting algorithm that effectively reduces local coverage gaps.
Abstract (translated)
URL
https://arxiv.org/abs/2610.29927
PDF
-
CausalAlign: Causal Alignment of Video Generation Models for Consistent Ego-Motion
2026-10-21 15:54:57Kai Zhu, Yintianrun Zhang, Shuwei Jiang, Ruolin Zuo, Yixin Chen, Duomin Wang, Miao Liu, Yin Li, Daquan Zhou, Yiwu ZhongAbstract
Diffusion-based video generation models (VGMs) can produce visually plausible videos, yet their camera ego-motion often drifts over long horizons. Camera ego-motion determines how the viewpoint changes between frames and is essential for maintaining consistent geometry and trajectories. To address this problem, we propose CausalAlign, a causal alignment framework with two components. First, we diagnose how precise ego motion is represented, used, and retained within VGMs using linear probes and causal interventions, revealing a rise-and-fall pattern during denoising. Second, we strengthen ego-motion state in VGM by aligning the representations at the identified layers to a 3D foundation model through delta-token space. This space encodes the changes between consecutive frames, allowing the alignment to focus on ego-motion dynamics rather than static scene content. With Wan2.2 fine-tuned on DL3DV and RealEstate10K, CausalAlign can largely improve motion accuracy, geometric consistency, and temporal smoothness, while preserving visual quality. For example, it improves geometric consistency by more than 20% relative to the base model. Moreover, with the fine-tuned VGM, CausalAlign improves the success rate of world action models (WAMs) by 10 points on mobile manipulation tasks of RoboCasa365 benchmark. Overall, CausalAlign improves ego-motion consistency in diffusion-based VGMs and yields noticeable gains in embodied models.
Abstract (translated)
URL
https://arxiv.org/abs/2610.29915
PDF
-
Automated Materials Data and Knowledge Extraction using Machine Learning and Large Language Models
2026-10-21 15:38:05Aravindan Kamatchi Sundaram, B. Pabitramohan Prusty, Rohit BatraAbstract
Over the past decade, substantial efforts have been devoted to developing machine learning (ML) methods for extracting materials data and, where possible, materials knowledge directly from the scientific literature. This review traces the evolution of these methods from heuristic, rule-based pipelines through natural language processing (NLP) models and transformer-based encoders to, lastly, large language models (LLMs) and agentic systems capable of extracting materials information across multiple modalities, including text, tables, and figures. Beside construction of large, queryable materials databases and knowledge graphs that support downstream materials discovery, we discuss how these advances have enabled new capabilities, such as, generic question answering across materials science, AI co-scientists and chatbots for research assistance, and closed-loop optimization and discovery when integrated with autonomous laboratories. We critically assess the datasets generated using these approaches, their extraction accuracy and computational costs, and the respective strengths and limitations of different methodologies, and even provide practical guidelines for developing more reliable and scalable materials data extraction pipelines. Finally, we identify key open challenges, including the extraction of full materials structure-property-processing-performance tetrahedron at the sample level, the scarcity of reliable benchmark datasets for systematic evaluation, and the resolution of non-standard materials terminology across different document modalities.
Abstract (translated)
URL
https://arxiv.org/abs/2610.29895
PDF
-
GeoMem-VLA: Internalizing 3D Scene Understanding into Temporal Memory for Vision-Language-Action Models
2026-10-21 15:23:18Qinfan Zhang, Yifan Luo, Haoyuan Shi, Guanlin Liu, Han Dong, Haozhe Shan, Yong Dai, Xiangcong Ren, Xiaozhu JuAbstract
Geometric structure is essential for robotic manipulation and understanding the 3D world. Most VLA models rely on 2D representations of current observations, while geometry-aware approaches often require additional 3D information. However, geometry inferred independently from individual observations is often discarded rather than accumulated into a persistent task representation. Our key insight is that action generation needs a compact, task-relevant geometric abstraction that can be continuously updated, rather than a complete, renderable 3D reconstruction at every step. We introduce GeoMem-VLA, a VLA framework that internalizes explicit 3D supervision into dedicated geometric tokens and maintains them as temporal memory for action generation, without directly aligning the VLM's native visual embeddings with the teacher feature space. During training, a frozen geometry foundation model supervises learnable GeoPlan tokens through a geometric flow-matching objective over the current scene point map and future end-effector tracks. These tokens are integrated with the memory of observation representations, where historical information is retrieved, adaptively fused with current observations, and consolidated to retain event frames as the task progresses. At inference, the policy retrieves historical representations and fuses them with the current working representation to predict actions, without invoking the 3D teacher or explicitly reconstructing the scene. After benchmark-specific downstream adaptation, GeoMem-VLA achieves 98.9% average success on LIBERO and 91.10% / 91.24% on RoboTwin's clean/randomized settings. Across five real-world tasks, GeoMem-VLA achieves 78.0% average success. These results suggest that internalizing and remembering 3D scene understanding is a promising approach to robot control.
Abstract (translated)
URL
https://arxiv.org/abs/2610.29866
PDF
-
Robot Generalization Beyond Volume: The Diversity Path via Zero-Shot Capabilities
Abstract
General-purpose robots must handle tasks, objects and environments they were never trained on, yet zero-shot generalization remains one of the least solved problems in physical AI. In digital AI, zero-shot capability grew with the volume and, above all, the diversity of training data, much of it sourced from the crowd. Robotics instead relies on paid collection by a few well-funded teams, which yields volume without diversity. We formulate zero-shot generalization in physical AI as coverage of the deployment distribution, and show that more samples from the same collectors shrink the unseen region only slowly, while each new contributor can extend it. Building on this and on zero-shot learning, we propose a shared latent space for robot actions that combines multi-modal alignment, generative augmentation, and noisy-label learning. As a scalable source of diverse data, we propose a community of developers building applications on one standard, approachable robot, Origami. We close with a testable hypothesis and an evaluation protocol relating real-robot zero-shot success to the diversity of contributed data.
Abstract (translated)
URL
https://arxiv.org/abs/2610.29829
PDF
-
UserSWE: Evaluating Coding Agents in Multi-Turn Conversations with Dynamic Users
2026-10-21 14:46:22Zhan-Yu Wu, Ji Luo, Alyssa Wu, Junjielong Xu, Richeng Xuan, Zhichao Hu, Chongming Gao, Xinting HuAbstract
As coding agents become more capable, developers increasingly use them to fix bugs through multi-turn interactions. Existing benchmarks do not adequately cover this scenario. Most present a complete issue report in a single turn without interaction. Their evaluation can produce both false positives and false negatives, accepting incorrect patches or rejecting correct ones. We introduce UserSWE for evaluating coding agents' ability to fix repository-level bugs through multi-turn interactions with dynamic users and execution-grounded verdicts at every turn. Its construction pipeline turns repository history and existing single-turn benchmarks into multi-turn tasks. At each turn, information pools constrain what the simulated user can reveal, while a persona chain sampled from a transition model calibrated on 20,859 development sessions governs how it speaks. Tests run after every turn, and a pair of agents from different model families writes and gates each verdict, using test results as primary evidence and model judgment only for aspects the tests cannot resolve. We evaluate eleven agents on 72 tasks from 42 repositories and find that agent rankings change with the interaction budget. Claude Fable 5.1 resolves the most tasks in one turn, GPT-6-Astra resolves the most by the eighth turn, and Claude Opus 5 leads at most intermediate budgets. These results indicate that single-turn performance is insufficient to characterize interactive coding ability.
Abstract (translated)
URL
https://arxiv.org/abs/2610.29822
PDF
-
COMET: A coordinated multi-layer runtime monitoring framework for human-supervised AI-enabled safety systems
2026-10-21 14:40:34Niclas Flehmig, Mary Ann Lundteigen, Shen YinAbstract
Incorporating AI systems into safety systems requires monitoring to avoid silent faults and ensure safe and reliable operation. Single monitors often do not provide actionable information for human supervisors, who act as the system's adaptive controller. We present COMET, a coordinated multi-layer runtime monitoring framework for human-supervised AI-enabled safety systems in which each monitor is anchored in a different fault mode. Monitor outputs are combined in a coordination layer and mapped by a communication layer to a traffic-light display. In this work, COMET consists of three monitoring layers: signal integrity, physics consistency, and the AI system's competence envelope. The framework is evaluated on an open dataset of compressed-air leaks, where an AI system detects leaks from acoustic emissions, with held-out conditions simulating covariate shift, semantic shift, and sensor faults, events expected during runtime but unknown in their exact appearance. All monitors are frozen after development and tested zero-shot, and the framework is compared against no-monitoring and single-monitor setups. The bare AI system fails silently and confidently under all three shift types, whereas COMET reduces the trusted-wrong-answer rate to zero on covariate-shifted no-leak data and on three of four sensor faults, at a nominal false-alarm rate of $\approx 1\%$. Single monitors come close in detection but cannot attribute the fault mode. COMET converts their anonymous detection flags into diagnoses. Residual silent misses under semantic shift define the boundary of passive monitoring. Runtime monitoring frameworks for AI-enabled safety systems must therefore combine quantitative fault detection and isolation with qualitative, human-oriented feedback.
Abstract (translated)
URL
https://arxiv.org/abs/2610.29816
PDF
-
Painting Change Map as guidance for generating painting processes
2026-10-21 14:35:17Markus Pobitzer, Chang Liu, Niccol\`o Biondi, Nicu SebeAbstract
Digital art generation has undergone a massive paradigm shift, moving from static image synthesis to dynamic video generation that can mimic the human painting process. While existing models can synthesize visually stunning time-lapses of paintings being created or "unpainted," they largely treat the creative process as a black box. Current conditioning mechanisms are either too rigid, relying on static geometric cues like Canny edges or depth maps, or too holistic, utilizing simple scalar progress variables that offer no localized control. Consequently, users lack the agency to steer how a painting evolves over time. In this work, we argue that true control over the video generation of artwork requires a joint understanding of two fundamental dimensions: where an artist acts, and when those actions take place. To achieve this, we introduce the Painting Change Map, a novel spatio-temporal control mechanism that explicitly tracks normalized structural changes across a 1,000-timestep continuum, a resolution sufficient to capture fine-grained stroke-level changes while remaining computationally tractable. By mapping localized visual differences into an encoded RGB space, our representation enables fine-grained guidance for both the forward painting and reverse unpainting processes. Extensive benchmarking demonstrates the efficacy of our approach. We outperform existing control baselines and achieve competitive or superior results compared to state-of-the-art painting methods. Furthermore, under oracle conditions, our method surpasses all baselines by a significant margin. Code is available at this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2610.25369
PDF
-
Learning What to Remember: Resource-Efficient Self-Improvement for Lifelong Agents
2026-10-21 14:25:06Guannan Lai, Zi-Yang Bai, Jun-Peng Jiang, Han-Jia YeAbstract
Self-improving agents aim to continuously improve their behavior by learning from experience after deployment. However, accumulating and reusing more experience does not necessarily lead to better decisions, particularly when generation, interaction, and memory resources are jointly constrained. Under such stringent constraints, we find that several existing lifelong-memory methods provide limited benefits over simple retrieval or even no-memory agents. This reveals a fundamental challenge for resource-efficient self-improvement: agents must learn not only which past experiences are relevant, but which are actually worth reusing to improve future decisions. When memory capacity and opportunities for reasoning and recovery are limited, retrieving an ineffective experience can consume scarce resources and hinder downstream execution. To address this challenge, we propose TokenWorth, a lightweight feedback-driven memory adaptation framework that learns query-dependent memory worth from sparse execution feedback. Without updating the backbone model, TokenWorth uses the learned worth to guide both experience retrieval and bounded memory management, enabling the agent to continually adapt how it reuses accumulated experience. Across three LifelongAgentBench environments and both local and API-based deployments, TokenWorth achieves the highest task success rate while using the fewest output tokens among the evaluated methods under the specified resource budgets. These results suggest that effective self-improvement under resource constraints depends not on accumulating more experience, but on learning what is worth remembering. Our code is publicly available at this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2610.26453
PDF
