Performance
TileBench: A Controlled Benchmark for Performance Evaluation and Bottleneck Diagnosis of Tile-Based Programming Models
Bowen Cui, Zhongchun Zhou, Hao Wu et al. (+4 more)
Tile-based programming models, such as Triton and cuTile, aim to simplify high-performance kernel development, but their practical performance, tuning behavior, and usability remain difficult to compare systematically. We present TileBench, a controlled
benchmark for evaluating Triton and cuTile on NVIDIA B200 GPUs under matched operator semantics and comparable implementation structures. TileBench contains 45 operators covering diverse AI-kernel patterns and
memory/computation behaviors. Each task provides a PyTorch reference, verified Triton and cuTile implementations, standardized data-types (dtype) and input-size sweeps, default and autotuned configurations, roofline-based metrics, and profiling-guided diagnosis. Our evaluation shows that performance gaps are workload-dependent: cuTile excels on a small cluster of Tensor-Core/TMA-friendly kernels, while Triton is stronger on many irregular, streaming, and bandwidth-bound operators. We further evaluate LLM-generated cuTile and Triton kernels and find that Triton is consistently more
token-efficient than cuTile under the same iterative refinement protocol. TileBench is publicly available at https://github.com/Deep-Learning-Profiling-Tools/Tilebench.
Read the full abstract ▾
All authors Bowen Cui, Zhongchun Zhou, Hao Wu, Tejas Ramesh, Junyu Yin, Jialiang Gu, Keren Zhou
Robotics
DA-GRD: Decision-Aware Grasp-Relevant Disambiguation for tactile recovery under perception-to-execution mismatches
Haoran Wang, Yuteng Sun, Yuanjie Li et al. (+5 more)
Grasping is a fundamental robotic capability that bridges perception and physical task execution. This paper studies grasp pose recovery under a perception-to-execution mismatch, where a grasp generated from visual perception may become spatially stale if the object moves before execution, using only sparse tactile interactions and no further visual observations. We propose DA-GRD, Decision-Aware Grasp-Relevant Disambiguation, which maintains a weighted planar belief over possible object configurations and selects tactile probes according to their ability to eliminate hypotheses and improve agreement among candidate task grasps. Rather than fully relocalizing the object, DA-GRD stops when the remaining hypotheses support a common executable grasp. In MuJoCo experiments on ten rigid objects with translations up to 5~cm and yaw perturbations up to $\pm45^\circ$, DA-GRD achieves an 84.7% physical lift success rate, compared with 9.1% for stale AnyGrasp, 21.2% for the original fix-scan baseline, and 63.7% for fix-scan method adapted with an SE(2) belief. DA-GRD also achieves a 57.3% Task conditioned Success rate. Across objects, it uses a success-average of 4.13 tactile probes over the ten per-object means, corresponding to a 72.5% reduction relative to the fixed 15-probe baselines. Real-world experiments on six objects achieve 71.7% physical lift success and 38.3% task-conditioned success with 4.20 probes on average. These results show that tactile sensing can recover task-relevant grasps under
vision-off conditions with limited physical interaction, without requiring complete object localization.
Read the full abstract ▾
All authors Haoran Wang, Yuteng Sun, Yuanjie Li, Ruofei Bai, Meng Yee Chuah, Wenyu Liang, Jun Li, Wei-Yun Yau
Computer Vision
EIB-Net: Entropy-Guided Information Bottleneck for Generalizable AI-Generated Image Detection
Zhida Zhang, Xinlei Ma and Jie Cao
The proliferation of photorealistic AI-generated images demands robust detection methods that generalize across diverse generative models. While existing approaches target manipulation-based forgeries with local artifacts, generation-based images (e.g., from
diffusion models) lack such traces, posing a fundamental challenge. We observe that generative models prioritize global semantics at the expense of local texture fidelity, making low-texture regions key indicators of synthetic origin. To exploit this, we propose EIB-Net, an
Entropy-guided Information Bottleneck Network. EIB-Net introduces a novel Image
Entropy (IE) metric to automatically select the most informative (lowest-
entropy) patch, then processes it with a Variational Information Bottleneck (VIB) to learn compact, generalizable features. Extensive experiments on DIFF, DiffusionForensics, and GenImage
benchmarks demonstrate state-of-the-art performance: EIB-Net achieves 85.7% accuracy using only 2% of training data, outperforming full-image baselines by over 15%, and maintains robust cross-generator generalization (83.5% average accuracy on GenImage). Furthermore, our
entropy-guided patch selection (EGPL) consistently enhances diverse backbones (CNNs and
Transformers), proving its practical value for data-efficient detection.
Read the full abstract ▾
Human-Computer Interaction
A Procedure for Classifying Attachments and Affective Social Bonds in Human-Robot Dyads
Imran Khan and Elisabeth Blagrove
Human-
robot interaction (HRI) claims that people form attachments and social bonds with artificial
agents, yet the terms are often applied without the behavioural and physiological criteria that give them content in their source disciplines. Without this empirical grounding, studies deploy widely divergent methods, frequently producing expansive relational claims that far outstrip their underlying evidence. To address this, we propose a standardised four-question procedure, grounded in criteria established in the developmental, ethological, and neuroendocrine literatures, that classifies a given human-
robot tie as an attachment, an affective social bond, or no relationship, with intermediate classifications when evidence is incomplete. We specify minimum evidential requirements for each question, and provide candidate HRI study designs, adapted from validated human-human, human-animal, and animal-animal paradigms. We then demonstrate the procedure by applying it to a representative set of published HRI studies, showing how often relational claims outstrip what the reported designs can establish. Finally, we discuss the ethical and regulatory burdens created when artificial
agents engage human biobehavioural systems. By replacing the divergent operationalisations with a unified, criterion-based classification, this paper gives HRI practitioners a standardised basis for evaluating, classifying, and comparing human-
robot relationships, and sets out the experimental rigour that each classification demands. We therefore call on researchers of human-
robot relationships to adopt such rigour, or to consider alternative terminology in their descriptions of these ties.
Read the full abstract ▾
Software Engineering
From Human Intent to Strategy Decisions: A Review Framework Integrating Large Language Models and Digital Twins
Teng-Hsien Ko and Chin-Te Lin
Automation systems must adapt to changing tasks, equipment states, and staffing conditions while providing evidence for human review. This study presents an intent-review framework integrating large
language models, digital twins, and human decision-making, implemented for multi-line task adjustment. A Propose-Verify-Decide workflow translates operator intent into structured requirements, generates a bounded set of candidate strategies, and checks semantics, simulation execution, and operational constraints. Linked records preserve traceability from requests to verification evidence and decisions. Thirty fixed test records were evaluated using four virtual
surgical-instrument sorting lines: 28 assessed the workflow and two assessed model generation. Eighteen workflow cases met expectations (64.29%); three of ten strategy-workflow requests completed autonomously (30%); and seven of eight invalid inputs were correctly intercepted (87.5%). Analysis by anomaly source showed that the framework rejected two candidate batches that failed throughput constraints and paused recommendations in two cases with incomplete evidence. A mistranslated priority rule nevertheless reached simulation. These findings demonstrate handling of specific invalid inputs, throughput violations, and evidence gaps, and identify consistency between original intent and execution specifications as a key improvement area. The effect on overall erroneous decisions remains to be evaluated through matched comparisons, together with the influence of digital-twin fidelity using independent physical references.
Read the full abstract ▾
Machine Learning · Cross-listed from stat.ML
Transformers as Cross-Task Learners: Shared Structure Drives Sample Efficiency in In-Context Learning
Zhongjie Shi, Rongjie Lai, Alexander Cloninger et al. (+1 more)
Transformers achieve remarkable performance by jointly learning broad families of tasks during pretraining and adapting to unseen tasks from only a short
prompt. Yet a rigorous mathematical and statistical understanding of this phenomenon remains limited. This paper aims to study how
Transformers exploit shared cross-task structure and how this structure affects the sample complexity of in-context learning (ICL). Specifically, we characterize task-space complexity through covering numbers under a prescribed metric, thereby quantifying the low-dimensional cross-task structure without requiring an explicit parametric representation. The resulting cover provides a set of anchor functions, which we use to introduce a task-identification-and-evaluation procedure: context observations localize an unseen task among the anchor functions, and the response at a query is predicted by aggregating the corresponding anchor function query evaluations. For approximation, we explicitly construct a
Transformer with Softmax
attention to approximate this procedure. For generalization, we derive an error bound that separates the effects of the number of pretraining tasks and the
prompt length. The scaling with respect to the number of pretraining tasks is governed by the intrinsic dimensions of the task space and input domain; once sufficiently many tasks are available, the dependence on the
prompt context length becomes dimension-free. To the best of our knowledge, this is the first work to quantify cross-task complexity for general nonlinear task families and explicitly construct a
Transformer that exploits their low-dimensional structure to perform ICL. Our theory provides a quantitative explanation of how joint pretraining across related tasks improves in-context generalization.
Read the full abstract ▾
All authors Zhongjie Shi, Rongjie Lai, Alexander Cloninger, Wenjing Liao
Computation and Language
Empath: Tracing Multi-Level Emotion Dynamics in Crisis Counseling Dialogues
Ziwei Gong, Yuchen Huang, Wen Liang et al. (+4 more)
Emotion dynamics are critical for understanding crisis-support conversations, yet most computational work treats
emotion as static utterance-level labels. We introduce EMPATH, a framework for understanding affective dynamics in mental health dialogues across three granularities: turn-level labels, transition probabilities, and global conversation archetypes. Applying EMPATH to text-based crisis conversations with self-identified Black texters discussing grief, we find persistent negative affect, gradual hope-ward transitions, distinct texter-volunteer emotional roles, and heterogeneous recovery trajectories. These results highlight the informative patterns that emerge from computationally understanding crisis support and expressions of grief as dynamic processes within conversations, as well as the overall value of
emotion-dynamic analysis for analyzing and comparing affect in dialogues.
Read the full abstract ▾
All authors Ziwei Gong, Yuchen Huang, Wen Liang, Nicholas Deas, Melanie Subbiah, Kathleen McKeown, Julia Hirschberg
Discrete Mathematics
Zero Forcing Sets in Temporal Graphs
Julien Baste, Simon Dreyer, Clara Marcille et al. (+2 more)
The Zero Forcing (or corruption) of a graph is the problem of finding a minimum-size “corrupting” set. It corresponds to a subset of its vertices that can corrupt the whole graph by iterating the following rule: if a corrupted vertex has exactly one neighbor that is not yet corrupted, the neighbor gets corrupted. The iteration of this process comes from the fact that the corruption of a vertex might enable new corruptions (from itself or some of its neighbors). For this reason, one can consider a step of corruption, where all the possible instances of the corruption rule are applied at once. This paper investigates Zero Forcing on temporal graphs, where the
topology of the graph evolves throughout the experiment. At each time step (or snapshot) of the graph, a step of corruption is resolved wherever possible. We study the problem of finding a minimum-size corrupting set such that the whole (temporal) graph is corrupted at the end of the experiment. We present a panorama of results, including NP-hardness in some not-so-restrictive scenarios,
polynomial algorithms, and a solution to an open question when the whole graph must be corrupted in a single step.
Read the full abstract ▾
All authors Julien Baste, Simon Dreyer, Clara Marcille, Mikaël Rabie, Ronan Toullec-Streicher
Artificial Intelligence
From Self-Distillation to Self-Practice: Privileged Information for Multi-Turn Agents
Xingyu Su, Abhishek Kumar, Qing Ping et al. (+5 more)
On-policy self-
distillation (OPSD) has become a popular recipe for post-training LLM
agents. It supervises the
agent model at the
token level with a stronger teacher view of the same model, obtained by conditioning on privileged information (PI). In this work, we show that in multi-turn
agents, this paradigm teaches the student to act with confidence but without the information behind it. The trained
agent behaves as if it had privileged information it never observed, and its performance falls well short of plain RL, in the worst case below the untrained base model. Therefore, we propose Privileged Self-Practice (PSP), which keeps the PI and moves it from the loss to the sampler. When the student's rollouts on a task mostly fail, we inject a short per-task instruction written by an analyzer model, sample the task again with the instruction in context, and train on the result with an unchanged GRPO objective. The privileged information stays in the
prompt and never enters the loss. Across AppWorld and SWE-bench Verified, with three different student models, PSP obtains the best average score in every setting and is the only method that consistently outperforms plain GRPO, with relative improvements over GRPO of 7% to 58% in average score across settings. Project page: https://sux1ngyu.github.io/PSP/.
Read the full abstract ▾
All authors Xingyu Su, Abhishek Kumar, Qing Ping, Youzhi Luo, Jonathan Buck, Zach Zhang, Subramanian Chidambaram, Vinayak Arannil
Artificial Intelligence
SLCA-GRPO: Resolving Cross-Segment Credit Misattribution in Tool-Calling RL
Yan Zhan, Shaobo Liu, Qiunan Liu et al. (+7 more)
Tool-calling
agents produce heterogeneous outputs, interleaving structured tool invocations with user-facing natural
language summaries. This output heterogeneity presents a structural failure mode in standard on-policy Reinforcement Learning (RL):
algorithms like GRPO indiscriminately broadcast a homogeneous trajectory-level scalar advantage to all
tokens. Consequently, gradient noise from summary generation leaks into tool-decision
tokens, causing cross-segment credit misattribution and brittle optimization. In this work, we propose SLCA-GRPO, a framework incorporating Segment-Locked Credit Assignment (SLCA). To enable scalable exploration without costly real APIs and stable training, we first construct the Schema-Guided LLM Simulator (SGLS) as foundational training infrastructure. Building on this, SLCA decouples advantage estimation at the structural segment level within a single group of rollouts, without requiring additional rollouts from intermediate states. Supported by Hierarchical Rewards (HierR), SLCA routes execution advantages to tool
tokens and preference advantages to summary
tokens, eliminating advantage contamination (the dominant cross-segment credit misattribution channel) within each policy update. On a 7B backbone, SLCA-GRPO accelerates convergence and outperforms standard GRPO, ToolPO, and RLTR by +2.53 pp on in-domain evaluation, +1.36 pp on the Berkeley Function-Calling Leaderboard (BFCL), and +9.15 pp on $\tau^2$-Bench under the same training budgets, achieving higher accuracy with reduced tool redundancy and costs.
Read the full abstract ▾
All authors Yan Zhan, Shaobo Liu, Qiunan Liu, Yuanjun Shi, Siqi Xu, WeiYi Hou, Xiang Xu, Zekang Li, Weizhou Pan, Jiahong Yan
Systems and Control
Bearing-Only Formation Tracking Control for Euler-Lagrange Multi-Agent Systems Without Inter-Agent Communication
Zilong Song, Lu Liu and Gang Feng
This paper investigates communication-free bearing-only formation tracking control for multi-
agent systems governed by Euler-Lagrange dynamics. Distinct from existing results that can only stabilize a stationary formation, this work considers a scenario where the leaders move with time-varying velocities while the inter-
agent communication is absent. In this setup, the leaders' states (position and velocity) are unavailable to all followers and cannot be estimated via distributed observers. A novel adaptive distributed control scheme is developed to address this problem. The design exploits the fact that bearing rates contain the projected relative-velocity information, which, together with bearing
rigidity, provides a
rigidity-based damping mechanism for compensating the unavailable velocity error. Moreover, this damping mechanism is incorporated into a bearing-driven auxiliary variable to construct a surrogate velocity error, facilitating the adaptive control design for EL dynamics. Furthermore, since this damping mechanism necessitates sufficient bearing
rigidity, we characterize a
rigidity-preserving set and establish its forward invariance, thereby guaranteeing such
rigidity via initial conditions. Via a Filippov-based Lyapunov analysis, the proposed scheme is shown to achieve local practical formation tracking in the sense that the velocity error converges to zero and the position error is uniformly ultimately bounded. As a corollary, for the constant-velocity case, asymptotic tracking is achieved without initial-condition restriction. The simulation results verify the effectiveness of the proposed control law.
Read the full abstract ▾
Computer Vision
Shamanthak Hegde, Xiangrui Liu, Maitreya Patel et al. (+1 more)
Unified
vision-
language models (VLMs) that tokenize images through a vector-quantized (VQ) codebook routinely hallucinate objects on grounded yes/no
benchmarks, yet existing decoding-time fixes treat this as generic miscalibration without an architectural account. Using activation patching across twenty-five models spanning eight LLM families, we identify an early-layer ($L_0$)
attention routing
circuit shared across VQ-tokenized VLMs and propose a three-gate diagnostic that distinguishes the models carrying it from those that do not. The diagnostic isolates ten positive models (five natural unified-VQ VLMs across three LLM families and five induced variants) and rejects the remaining fifteen. A single-variable architectural swap (LLaVA-1.6 CLIP+MLP $\rightarrow$ VQ+Linear) installs the
circuit, while a matched-compute MLP control on identical data does not, isolating vector quantization as the source of the pathological signal; the routing pathway that carries it is one that the backbone already provides. Against tuned VCD and DoLA baselines, tuned DoLA wins on binary calibration, but \textbf{only $L_0$ ablation reduces object
hallucination in open-ended generation} (CHAIR$_i$ reduces by $31\,\%$ relatively, whereas tuned DoLA and VCD leave it unchanged or worsen it). These results recast object
hallucination in unified VQ VLMs as a property of architecture and pretraining, and yield a targeted intervention that mechanism-agnostic decoding cannot replicate.
Read the full abstract ▾
All authors Shamanthak Hegde, Xiangrui Liu, Maitreya Patel, Yezhou Yang
Cryptography and Security
The Tokens Remember: When Tokenization Bypasses Knowledge Editing and Unlearning
Manit Baser, Aditya Nawal, Dinil Mon Divakaran et al. (+1 more)
Open-weight LLMs give downstream users control over the inference stack, but this flexibility can undermine post-release guarantees that sensitive knowledge has been modified or removed. Model editing and machine unlearning are used to modify or remove targeted knowledge without retraining models from scratch. However, existing security evaluations of these techniques face two critical limitations. First, they typically require access to either the original pre-edit/unlearning model or auxiliary classifiers to detect modifications or reconstruct pre-edit behavior. Second, they evaluate modifications under the canonical tokenization of an input, implicitly treating tokenization as a benign preprocessing step. We show that this assumption creates a security gap: the same input string can be represented by alternative valid tokenizations that induce different computational trajectories, allowing an adversary to bypass localized modifications and recover information intended to be suppressed. We introduce Toketive, a simple yet powerful reference-free attack that exploits the tokenization-based side channel to (i) detect modified knowledge and (ii) reconstruct the corresponding pre-edit response. It operates solely on the released model and requires neither the pre-edit model, training data, shadow models, nor auxiliary classifiers. Across five LLMs, six
datasets, and six editing and unlearning techniques, we find that 38.6% of alternative tokenizations bypass the modification and recover the pre-edit response. Toketive detects modified facts with an F1 score of 84.2%, a 26.2% relative gain over the strongest baseline, and reconstructs pre-edit responses with 74.5% top-5 accuracy, 21.7% higher than the best baseline. Our results show that localized modifications should not be treated as robust knowledge-control boundaries without
adversarial evaluation over alternative representations.
Read the full abstract ▾
All authors Manit Baser, Aditya Nawal, Dinil Mon Divakaran, Mohan Gurusamy
Information Theory
Multi-Agent Orchestration of 3GPP Channel Estimators
I. Zakir Ahmed and Hamid Sadjadpour
Pilot-aided channel estimation is a decisive block in orthogonal frequency-division multiplexing (OFDM) receivers for both 5G New
Radio (5G-NR) and Long-Term
Evolution (LTE). A large body of
estimators exists, from simple least-squares (LS) interpolation to statistically optimal linear minimum-mean-square-error (LMMSE) variants and, more recently, deep convolutional denoisers, yet no single
estimator is uniformly best: the winner depends on the propagation scenario, the numerology, the operating signal-to-noise ratio (SNR), the mobility (Doppler), and the
antenna configuration. In this paper, we quantify this fact through a unified study of eight literature
estimators evaluated over the 3GPP TR~38.901 Urban-Macro (UMa), Urban-Micro (UMi), and Rural-Macro (RMa) channels generated with NVIDIA Sionna, for both 5G-NR and LTE numerologies, in single-input single-output (SISO) and $8\times2$ multiple-input multiple-output (
MIMO) settings. We then propose a condition-adaptive multi-
agent orchestrator that treats each
estimator as an independent
agent and dispatches, per operating condition, to the
agent that is best on a validation split without any genie knowledge. The orchestrator tracks the per-realization oracle to within $1.07$~dB and improves the normalized mean-square error (NMSE) over the best fixed strategy by up to $3.6$~dB at high SNR, where the low-SNR champion is no longer optimal. Because the
agents are independent, running them concurrently delivers this best-of-eight accuracy at essentially single-
estimator latency: a data-parallel partition scales the wall-clock nearly as $1/K$ with $K$ workers (up to $6.9\times$), whereas naive by-
algorithm partitioning is Amdahl-limited by the heaviest
agent. The results substantiate multi-
agent orchestration as a practical route to robust channel estimation across heterogeneous 5G-NR/LTE deployments.
Read the full abstract ▾
Robotics
Design and Evaluation of LLM Chaining-Based Task Planning for General Purpose Service Robots
Lucas Da Mota Bruno, Jiahao Sim and Yoshinobu Hagiwara
General Purpose Service
Robot (GPSR) tasks, as defined in the RoboCup@Home
benchmark, require
robots to interpret diverse natural
language commands and generate multi-step action sequences in real home environments. Conventional Single
Prompt (SP) approaches suffer from context bloat and the "Lost in the Middle" phenomenon, leading to unreliable task planning. We propose an LLM chaining architecture that separates instruction classification and action generation into two specialized stages, reducing per-inference
prompt length by approximately 45% while improving planning consistency. We evaluate our method using 100 randomly generated GPSR commands across three
language models spanning local open-source and frontier cloud deployment contexts. Results show consistent planning improvements over SP across all models, with gains of up to +37 percentage points on local models. Further, real-
robot execution experiments on the Toyota Human Support
Robot (HSR) reveal that planning success alone does not guarantee task completion, with 6 of 10 tasks completing successfully and execution-layer failures identified as the primary remaining bottleneck.
Read the full abstract ▾
Multimedia
From Scattered Gaussians to Structured Maps: Efficient Gaussian Splatting Coding via Dual-phase Morton Sorting
Bolin Chen, Shanzhi Yin, Ru-Ling Liao et al. (+2 more)
3D Gaussian Splatting (3DGS) enables high fidelity novel view synthesis but suffers from excessive storage and bandwidth requirements due to its unstructured representation. To address this, a projection based
video coding framework has emerged as a leading approach, supported by MPEG's ongoing standardization, where 3DGS attributes are converted into 2D maps to take advantage of efficient compression using established
video codecs such as HEVC and VVC. However, the effectiveness of this approach depends heavily on the spatial coherence of the projected
video, which current sorting strategies such as PLAS and Morton ordering fail to preserve adequately, either incurring high computational cost or achieving limited correlation retention. To overcome these limitations, we propose a dual phase Morton spatial sorting
algorithm that improves both coding efficiency and processing speed. In the first phase, Morton based 1D indexing is applied to high dimensional attributes to enhance spatial locality. The second phase further refines layout continuity through a structured 2D Morton mapping table that enforces spatial adjacency. This hierarchical strategy generates highly regular, block wise feature maps with strong local correlation, making them well suited for compression via conventional block based coding tools. Experimental results show that our method significantly outperforms existing approaches in both compression performance and runtime efficiency, providing a practical and standard compatible solution for 3DGS data coding.
Read the full abstract ▾
All authors Bolin Chen, Shanzhi Yin, Ru-Ling Liao, Yibo Fan, Yan Ye
Sound
The Vulnerability of Neural Audio Watermarks under Speech Enhancement
Xincong Zhong, Shengyao Wang, Lingfeng Yao et al. (+4 more)
Neural audio watermarks are increasingly deployed in commercial
speech generation systems to make AI-generated
speech traceable, yet their robustness has been studied mainly under conventional signal distortions. Since a watermark can be regarded as imperceptible noise added to the
speech signal, a natural question is whether
speech enhancement (SE), as a denoising model, can remove it. In this paper, we cascade Gaussian noise with SE models as a black-box watermark removal attack, covering both discriminative and generative SE paradigms, against six neural watermarks: AudioSeal, WavMark, SilentCipher, Timbre, Perth, and AlignMark. Experimental results show that the proposed attack significantly outperforms existing neural re-synthesis methods in watermark removal. In particular, we find that generative SE, which reconstructs the harmonic regions of
speech while denoising, is highly destructive to watermarks. These findings show that SE poses a serious threat to current audio watermarking methods, and we call for SE-aware robustness evaluation in watermark design.
Read the full abstract ▾
All authors Xincong Zhong, Shengyao Wang, Lingfeng Yao, Yihang Bao, Jinze Yu, Miao Pan, Jiang Liu
Machine Learning · Cross-listed from stat.ML
Personalised federated learning for Riemannian and Euclidean EEG decoding
Thibault Pautrel, Florent Bouchard, Ammar Mian et al. (+1 more)
Federated learning (FL) lets
EEG decoders learn from recordings of several subjects without pooling them. We consider two light
EEG decoders, the Riemannian SPDNet and the Euclidean EEGNet. Both split into a trunk, which builds a latent representation, and a head, which classifies it. Inter-subject variability, however, makes a single shared FL model a poor fit for each subject. Personalised FL addresses this: all subjects learn a common trunk, and each subject keeps its own head. We adapt it for SPDNet and study its effects against standard FL and centralised training, with EEGNet as a Euclidean baseline. Experiments cover three motor-imagery
datasets that span diverse regimes in channels, subjects and classes. We observe that personalised SPDNet reaches higher accuracy than both standard FL and centralised training, while converging in fewer rounds and communicating fewer parameters than standard FL. It also outperforms every EEGNet configuration on two of the three
datasets, although centralised EEGNet outperforms centralised SPDNet.
Read the full abstract ▾
All authors Thibault Pautrel, Florent Bouchard, Ammar Mian, Guillaume Ginolhac
Performance
Paging the Experts: A Reproducible Characterization of Flash-Backed MoE Inference on iPhone
Musa Shams
Sparse activation reduces mixture-of-experts computation without eliminating the need to store all experts. We present Routide, a Swift/MLX runtime that executes the text path of a pinned public Qwen3.6-35B-A3B quantized checkpoint while keeping expert weights in iPhone storage and a byte-budgeted subset in
memory. We characterize cache-policy sensitivity, numerical comparison boundaries, and measurement limits. Across five recorded 128-
token workloads, fixed-route replay gives 0.00% demand hits with a 512 MiB LRU cache, 18.80% with seeded random eviction at the same budget, and 38.58% with 576 MiB LRU. The apparent capacity cliff is therefore a policy/workload interaction, not a universal
memory requirement. Same-runtime Mac controls preserve generated sequences across eviction and asynchronous prefetch, including 2,560 exact
token comparisons and 10,334 speculative loads. In contrast, complete resident-Python versus recorded-phone sequences disagree on all five tested cases, precluding a general numerical equivalence claim. Two separately scoped iOS 27
memory protocols observe sampled process-footprint peaks of 1.87-2.32 GiB on short
prompts and 2.39-2.73 GiB on one longer
prompt. We retain a thermal stopping event, negative timing comparisons, and a single qualified whole-device power estimate. These results establish bounded feasibility and identify limitations that a deployment claim must not hide.
Read the full abstract ▾
Robotics
Simple Torque-Observation Alignment for Zero-Shot Sim-to-Real Grasping with a Direct-Drive Gripper
Doyoung Kim, Edgar Lee, Hyeonsun Park et al. (+4 more)
Torque observations in reinforcement learning remain challenging because simulated and measured torque differ in scale, offset, and noise. In this paper, we propose a simple torque observation
alignment method for
robots with direct-drive (DD) actuators, in which motor current maps linearly to joint torque through a motor-type-specific torque constant K_tau. First, dynamometer calibration identifies K_tau* and corrects the scale mismatch between simulated and real torque. Second, the method uses delta_tau(t) = tau(t) - tau(t-1) as the observation in both domains to eliminate the constant offset instead of using the direct torque tau(t), which carries a domain-dependent bias. Third, Gaussian noise obtained from the dynamometer measurement data is injected during the learning process. To validate the proposed method, we train a teacher-student grasping policy entirely in simulation and deploy the distilled student on a multifingered DD gripper. The deployed policy performs proprioceptive grasping using only joint positions and torque differences. We conduct an ablation study comparing the proposed method with alternative
alignment variants on nine in-distribution (ID) objects. The proposed method achieves 100% grasp success. These results demonstrate that the proposed
alignment method improves the robustness of zero-shot policy transfer on the DD gripper against real-world torque-observation mismatches.
Read the full abstract ▾
All authors Doyoung Kim, Edgar Lee, Hyeonsun Park, Chunghyeon Lee, Chihyun Han, Uisu Hwang, Seokhwan Jeong
Computer Vision
Exploiting answer-invariant redundancies in satellite imagery for efficient VLM inference on edge
Ishani Janveja, Davis Zhang, Seoyul Oh et al. (+1 more)
Onboard
vision-
language models could enable satellites to answer queries directly, but exhaustive tiled inference over high-resolution imagery is slow and energy-intensive. We identify answer-invariant
token redundancy (AITR): image tiles and
vision tokens that can be removed without changing the final answer. We present Rift, a two-stage system that performs query-conditioned tile
pruning followed by elastic prefill to reduce
token budget. We evaluate it on LLaVA-1.5 7B running on Jetson AGX Orin. Compared with exhaustive tiled inference, Rift reduces energy by 78% and latency by 69%, while increasing accuracy from 45% to 73%.
Read the full abstract ▾
All authors Ishani Janveja, Davis Zhang, Seoyul Oh, Deepak Vasisht
Computer Vision
RGBD20K: A Large-Scale Benchmark for RGB-D Semantic Segmentation
Shaohua Dong, Zexuan Meng, Haiyan Sun et al. (+6 more)
In this paper, we propose RGBD20K, a novel
dataset for facilitating the development of more robust and general RGB-D semantic
segmentation by encompassing abundant categories and high-quality annotations. RGBD20K possesses several attractive properties: (1) Expanded Semantic Space. In particular, it covers 160 fine-grained categories, largely surpassing the category diversity of existing popular RGB-D
benchmarks (e.g., NYUv2 with 40 classes and SUN RGB-D with 37 classes). With such enriched semantic coverage, we expect to promote the learning of more generalizable
segmentation models. (2) Larger Scale. Compared with current
benchmarks, RGBD20K offers 20,000 RGB-D image pairs, providing a substantially larger training resource that benefits the development of more powerful deep models. (3) High-Fidelity Annotation. We perform rigorous re-evaluation and correction of existing labels to resolve long-standing annotation noise, resulting in a clean and reliable ground-truth foundation. Furthermore, we propose a novel score-purified fusion (SPF) method, which achieves state-of-the-art performance across all evaluated
benchmarks, demonstrating the effectiveness of our approach in leveraging high-quality
multimodal information for RGB-D semantic
segmentation. The
dataset is here: https://github.com/ShaohuaDong2021/RGBD20K/.
Read the full abstract ▾
All authors Shaohua Dong, Zexuan Meng, Haiyan Sun, Bing Fan, Cuicui Zhang, Dylan Joseph, Kewei Sha, Yunhe Feng, Heng Fan
Machine Learning
Generative Atmospheric Super-Resolution from Heterogeneous In Situ Observations through Composable Interfaces
Yang Xu, Dibyajyoti Chakraborty, Haiwen Guan et al. (+2 more)
Atmospheric observations are sparse, heterogeneous, and unevenly distributed, whereas many generative atmospheric models learn distributions over regularly gridded multivariate states. Once pretrained,
diffusion models can supply atmospheric priors that can be combined with observation-derived likelihood factors in a
Bayesian formulation. However, these observation sources differ substantially in geometry and sampling density, complicating the consistent use of their observations within a common inference framework. Here, we formulate this reconstruction problem as generative atmospheric super-resolution and introduce composable observation interfaces for conditioning a single pretrained 13-variable atmospheric
diffusion model. The interfaces convert sparse radiosonde (R), clustered aircraft (A), and dense irregular surface-station (S) observations into source-specific likelihood factors that specify where observations constrain the gridded state, how residuals are counted under uneven sampling, and how strongly each source guides posterior sampling. We developed the aircraft and surface observation interfaces using 2019 observations and evaluated the selected interfaces throughout 2020 without further tuning. Compared with reconstructions conditioned only on radiosonde observations, the composed R+A+S interface reduces RMSE evaluated against ERA5 by $9.24\%$ across all 13 state variables over the CONUS domain. The aircraft and surface factors provide complementary improvements in upper-air and surface variables. The R+A+S combination also lowers the Continuous Ranked Probability Score (CRPS), while evaluations at held-out aircraft and surface-station observations show reduced prediction errors. Together, these results demonstrate a modular route for conditioning a pretrained atmospheric generative prior on heterogeneous in situ observations without retraining the underlying model.
Read the full abstract ▾
All authors Yang Xu, Dibyajyoti Chakraborty, Haiwen Guan, Sen Wang, Romit Maulik
Machine Learning
Growth-Inspired Graph Generation and Inverse Design of Mechanical Lattices via Dot Matrices Database Augmentation and GCNN
Weiyun Xu and Jiamu Liu
Natural load-bearing and transport networks are not assembled in a single step; they emerge through a temporally ordered process of growth, branching, reinforcement, and loop formation. Inspired by this developmental logic, this work introduces a morphogenetic graph-generation framework for mechanical
lattices in which a discrete dot
matrix provides potential nodes and the final architecture is created by sequential cross-layer and intra-layer growth. The same rule is visualized in two dimensions as a leaf-vein-like developmental sequence and implemented in three dimensions on a 3x3x3 nodal
matrix containing 27 candidate nodes. A
dataset of distinct three-dimensional
lattices was evaluated by beam-based finite element analysis and represented directly as graphs. A graph convolutional neural network (GCNN) with three graph-convolution layers and dual global pooling learns the
topology-property mapping and predicts effective compressive stiffness. Coupling the GCNN surrogate with rapid structural sampling enables inverse design: for a target stiffness of 1000 MPa, the selected design was predicted at 1042.43 MPa and validated by finite element analysis at 1027.49 MPa. Beyond straight members, the framework has also been extended to parameterized horseshoe-shaped curved beams made of nonlinear materials, enabling
topology-geometry design toward prescribed deformation shapes. Our work provides a paradigm for augmenting the database of mechanical metamaterials, and the resulting perspective links biological morphogenesis, graph learning, and nonlinear shape programming in a unified generative design framework for architected materials.
Read the full abstract ▾
Robotics
CAMP: Cooperative Arm-Hand Motion Planning in Constrained Spaces
Ziyuan Wang, Yunlong Shan, Fei Mo et al. (+6 more)
Coordinated arm-hand motion planning is fundamental to dexterous robotic manipulation in complex and constrained environments. A straightforward solution is to decompose the problem into separate arm path planning and hand motion generation; however, this poses a dilemma: decomposition can miss feasible solutions that require coordinated arm-hand adaptation along the path. Alternatively, directly planning in the high-dimensional joint arm-hand configuration space captures such coupling but faces a substantially enlarged search space and nonconvex collision constraints. To characterize this coupling, we formulate feasible hand fibers that capture collision-free hand configurations for each arm configuration. Based on this formulation, we propose CAMP, a high-success and efficient cooperative arm-hand motion planner for constrained environments. CAMP constructs candidate trajectories through layered hand search with local arm relaxation, then compactly represents them using endpoint-preserving via-point movement primitives (VMPs) for coarse-to-fine joint optimization. Across six constrained simulation tasks, CAMP achieves 84.2-98.5% planning success, outperforming alternative planners with competitive efficiency. Ablation studies verify the contributions of arm relaxation, VMP representation, and coarse-to-fine optimization, while real-
robot experiments demonstrate CAMP on constrained manipulation tasks. The project website is available at https://camp-armhand.github.io/.
Read the full abstract ▾
All authors Ziyuan Wang, Yunlong Shan, Fei Mo, Sichao Liu, David Navarro-Alarcon, Jia Pan, Kosta Jovanovic, Xin Jiang, Peng Zhou
Robotics
Outcome-Sensitive Motion Search for Impact-Aware Dexterous Catching
Guorui Pei, Jinsong Wu, Songyuan Su et al. (+5 more)
Skilled humans can catch fast-moving objects softly by coordinating interception, velocity matching, and follow-through to mitigate impact. Learning such impact-aware catching with reinforcement learning (RL), however, is challenging, as the policy must achieve reliable interception and grasping while regulating the sensitive transition into contact. Moreover, even a capable privileged-state RL teacher may not provide ideal demonstrations for a deployable imitation-learning (IL) student: teacher failures limit task coverage, while small variations in pre-contact motion can produce substantially different impact and grasping outcomes. We characterize this phenomenon through interventional outcome sensitivity and introduce the outcome-sensitive window (OSW) to guide targeted demonstration construction. Building on this formulation, we propose Outcome-Sensitive Motion Search, which learns a task-conditioned
manifold of successful OSW motions and performs local geodesic search to refine successful teacher rollouts and repair task conditions where the teacher fails. We then validate candidate motions through complete rollouts under a calibrated IL-student action-error model and retain only successful executions as demonstrations. Extensive simulation experiments demonstrate that our method effectively repairs task conditions where the teacher fails and enables the resulting IL policy to outperform the privileged RL teacher in both catching success and impact mitigation.
Read the full abstract ▾
All authors Guorui Pei, Jinsong Wu, Songyuan Su, Jiaming Qi, Sichao Liu, David Navarro-Alarcon, Bin Liu, Peng Zhou
Robotics
CALM: Current Aligned Link Manipulation for Single Arm Oversized Object Lifting
Jun Hu, Sihan Chen, Kosta Jovanovic et al. (+4 more)
Most
robots manipulate objects solely with their end effectors, whereas humans flexibly leverage different body parts, such as the forearm and elbow, especially when handling oversized objects. Learning such whole-arm manipulation is chal-lenging due to long-horizon sparse rewards, limited contact sens-ing, and the sim-to-real gap in contact and actuator dynamics. To address these challenges, we propose Current-Aligned Link Manipulation, a framework for learning long-horizon contact-rich manipulation using motor current as joint load related feedback. Three stage-specific policies first learn repositioning, grasping, and lifting using privileged simulation information, and a stage router sequences them to generate complete task demonstrations. For sim-to-real transfer, a causal current mapper predicts physical motor current from simulated joint histories, aligning the actuator current observation between simulation and hardware. A unified student policy then learns from these demonstrations using only deployable sensor observations and is further refined with DAgger. The task policies are trained entirely in simulation, and the final student is deployed on hardware. Experiments demonstrate 76.2% (762/1000 trials) complete-task success in simulation and 73.3% success (22/30 trials) on the physical
robot for sequential oversized-object lifting.
Read the full abstract ▾
All authors Jun Hu, Sihan Chen, Kosta Jovanovic, David Navarro-Alarcon, Xueqian Wang, Jia Pan, Peng Zhou
Neural and Evolutionary Computing
EvoTreeNAD: Genealogy-Guided Evolution for LLM-Driven Neural Architecture Discovery
Lishan Yu, Derek Jiu, Qizhen Lan et al. (+1 more)
AI-driven scientific discovery accelerates research by autonomously developing solutions and designs. Large
language model (LLM)
agents support this process through iterative generation and evaluation. Yet these iterations alone do not ensure cumulative progress or establish which directions to pursue next. Costly evaluation further constrains the scope of exploration. Neural architecture discovery brings these challenges together, coupling open-ended design with resource-intensive experimentation. We introduce EvoTreeNAD, a genealogy-guided evolutionary
algorithm that constructs trainable architectures without a supplied seed or a hand-specified search space. Starting from an empty root, it grows a persistent genealogy in which each new node represents a complete architecture. Top-percentile values computed from each node and its descendants guide lineage selection. Using the selected design history, an Idea
Agent proposes a variant and a Code
Agent implements it. Each evaluated variant becomes a child node, expanding the genealogy while providing evidence for subsequent lineage selection. Our theoretical analysis establishes the existence of stationary variation regimes as the genealogy grows. Under specified variation assumptions, sustained top-percentile family values quantify the probability of generating high-reward architectures in these regimes. EvoTreeNAD discovers architectures that outperform the compared NAS and NAD baselines, achieving CIFAR-10/100 test errors of $2.05{\pm}0.06\%$ and $15.09{\pm}0.22\%$. On all six MedMNIST-v2 tasks, the discovered architectures surpass the strongest listed baselines. A controlled CIFAR-10 study further shows that EvoTreeNAD outperforms direct generation, best-of-$N$ greedy continuation, and full-family-mean routing.
Read the full abstract ▾
All authors Lishan Yu, Derek Jiu, Qizhen Lan, Xiaoqian Jiang
Artificial Intelligence
MeshHeal: Two-Timescale Self-Healing for Gray Failures in Decentralized LLM Agent Networks
Keru Chen, Sen Lin, Yingbin Liang et al. (+2 more)
Decentralized LLM-based multi-
agent systems coordinate through local interactions, but an
agent can remain responsive while its task-solving quality persistently degrades. Such gray failures require protecting current tasks before sufficient evidence exists to alter future routing, while still allowing recovered
agents to rejoin. We introduce MeshHeal, a fully decentralized self-healing framework that couples ability-matched peer review across two timescales. At the fast timescale, an adaptive hierarchy escalates uncertain or low-scoring outputs from repeated single-reviewer evaluation to committee deliberation and, when needed, correction before use. At the slow timescale, a task- and ability-conditioned peer-relative detector aggregates scores to distinguish persistent degradation from ordinary output variation, trigger mandatory committee review, and eventually exclude degraded
agents from ordinary routing; recovery probes provide fresh evidence for reintegration. To faithfully evaluate routing, we introduce Model-Backed MAS Evaluation, which ties ability assignments to execution models, since
prompt-based ability assignments alone can leave routing errors hidden. Across BBH, MATH, and MMLU-Pro, MeshHeal achieves 0.839 degraded-phase accuracy using 51k total model
tokens per task, versus the strongest baseline Symphony's 0.807 accuracy using 115k per task. Under staggered degradation and recovery, MeshHeal isolates degraded
agents, keeps them excluded from ordinary task execution until recovery, and returns them to normal routing.
Read the full abstract ▾
All authors Keru Chen, Sen Lin, Yingbin Liang, Nathaniel D. Bastian, Shaofeng Zou
Artificial Intelligence
AlphaDiverse: Post-Training Local Quantitative Research Agents for Diverse Exploration in Alpha Factor Mining
Qingzhuo Wang, Zikun Wei, Zhihua Wei et al. (+1 more)
Large
language model (LLM)-based multi-
agent systems can automate alpha factor mining, but their reliance on external APIs limits control over cost, availability, and confidentiality. Long research loops also tend to revisit a few successful economic mechanisms that lead to research path collapse. To address these limitations, we propose AlphaDiverse, a framework that integrates a multi-
agent alpha research system, diverse research path collection, and post-training for local
agents. We let the research system generate complementary plan portfolios and vary research environments across loops to collect diverse research paths. Using these diverse traces, we warm-start local Planner and Realizer
agents with supervised fine-tuning. Then, we propose a joint GRPO method to optimize both of them using predictive quality and diversity of contributions. Research feedback is confined to inner period data, while a frozen final model is evaluated on a later outer period data, thereby avoiding test-set tuning. Experiments across four Chinese stock universes show that AlphaDiverse can combine competitive prediction with broader exploration.
Read the full abstract ▾
All authors Qingzhuo Wang, Zikun Wei, Zhihua Wei, Wen Shen