Computation and Language
User Feedback Provides a Unique Signal that LLMs Can not Detect
Shachar Don-Yehiya, Leshem Choshen and Omri Abend
Harnessing naturally occurring feedback from user interactions offers a promising learning signal for Large
Language Models (LLMs). However, recent studies suggest this feedback is inherently noisy and difficult to leverage effectively. We challenge this conception by demonstrating that user feedback is a highly actionable signal for improvement, and that its perceived ineffectiveness stems from a systematic bias in current evaluation paradigms. To isolate the usefulness of feedback, we construct synthetic data with a definitive ground truth, alongside naturalistic data to validate that our findings hold in real-world scenarios. By comparing model revisions generated with and without access to feedback across both settings, we show that feedback-informed revisions resolve targeted issues at significantly higher rates than baseline revisions. Finally, we expose the root of the evaluation bias: when a model successfully fixes an issue exclusively due to feedback, LLM judges frequently fail to identify the genuinely corrected response, systematically preferring inferior baseline outputs instead.
Read the full abstract ▾
Machine Learning · Cross-listed from math.OC
Improved Gradient Descent Lower Bounds Beyond Nesterov
Yuhan Ye and Kaizhao Liu
We study how far gradient descent (GD) can be accelerated by predetermined stepsizes in smooth convex optimization. Going beyond the classical $\Omega(n^{-2})$ first-order oracle lower bound of Nemirovsky and Yudin (1983), we prove an $\Omega(n^{-1.6342})$ non-anytime lower bound and an $\Omega(n^{-1.2408})$ anytime lower bound. These improve the recent $\Omega(n^{-1.932})$ non-anytime lower bound of Ma and Chen (2026) and the $\Omega(n^{-4/3})$ anytime lower bound of Tsai et al. (2026), respectively. Both results continue to hold when the stepsizes may be negative. Our anytime lower bound also shows that the $O(n^{-\log_2(1+\sqrt{2})})$ rate of non-anytime silver schedules (Altschuler and Parrilo, 2025; Grimmer et al., 2025) is unattainable in the anytime setting. This establishes a strict separation between the two settings.
Read the full abstract ▾
Computer Vision
MuyBridge: Mobile Human Center-of-Mass Estimation from Monocular Video via Sparse Fusion
Aidan Bradshaw, Marco Giordano, David Rode et al. (+8 more)
The 3D center of mass (CoM) is a primary quantity in the biomechanical analysis of sport, rehabilitation, and clinical movement, yet existing 3D pose tracking, mesh recovery, and multi-view triangulation methods either optimize 3D keypoint accuracy without anatomical constraints or carry compute and capture infrastructure too heavy to deploy where CoM tracking is most useful. As a result, the metric CoM remains difficult for coaches and movement analysts to measure from a single camera where athletes train and compete. In this work, we introduce MuyBridge, an on-device system that estimates the athlete's segmental center of mass trajectory from a single phone camera
video stream. MuyBridge couples a compact 2D pose network and a distilled single-step
monocular depth network through an analytic metric fusion that uses anatomical and physical priors to anchor the metric CoM, requiring no 3D or task-specific supervision. Evaluated on the athletic movements of AthletePose3D (running, track and field, and figure skating), MuyBridge achieves 33-41 mm vertical CoM error and 2.3-6.6% absolute-relative range error (AbsRel) under a one-time calibration, and produces CoM estimates at the 63 FPS pose-estimation rate using asynchronous 2.86 Hz depth updates on iPhone 15. Code is available at: https://github.com/Abradshaw1/Muybridge
Read the full abstract ▾
All authors Aidan Bradshaw, Marco Giordano, David Rode, Andreas Habersack, Elif Basokur, Annika Kruse, Markus Tilp, Michele Magno, Peter Wolf, Luca Benini, Christoph Leitner
Machine Learning
The Implications of Linguistic Illegibility for LLM Security
James Mickens
LLMs are trained to generate natural
language. However, various strands of evidence indicate that an LLM's externalized linguistic outputs and mechanistically-extracted linguistic features can be an unreliable lens for understanding internal model computation. We introduce the term “linguistic illegibility” to broadly refer to scenarios in which an LLM's externalized or mechanistically-probed
language artifacts fail to represent how the model actually thinks. We argue that the specter of linguistic illegibility is unavoidable for LLMs whose internal computations are not directly expressed via
language, but rather math over activation spaces (with lossy translations between activation spaces and natural
language happening at the bookends). If linguistic illegibility is always possible, then security mechanisms that rely on a model's linguistic self-reporting (e.g., chain-of-thought monitoring, constitutional self-critique, activation probing for linguistically-defined feature vectors) can never be completely sound; the model sandbox will always need isolation techniques whose guarantees do not depend on reading a model's linguistic state at all. We argue that observing a model's outputs using taint tracking is a promising approach for an effective sandbox: regardless of how a model linguistically self-reports, a taint tracking policy can define, a priori, various pieces of system state that should never be influenced by model-produced data. We also discuss several additional sandboxing mechanisms (e.g., robust virtualization, third-party
auditing of sandboxing configurations) which collectively provide a critical floor beneath linguistic monitoring, and would have mitigated recent sandbox exploits by frontier models.
Read the full abstract ▾
Data Structures and Algorithms
Almost Linear 3-Spanners of Temporal Cliques
Julia Baligacs, Davide Bilò, Václav Blažej et al. (+2 more)
Temporal graphs model dynamic networks by assigning positive integer time labels to the edges, while information propagates along temporal paths, whose edge labels are traversed in nondecreasing order. A temporal $\alpha$-spanner of a temporal graph with $n$ vertices is a temporal subgraph that approximates the minimum-hop temporal distance between every pair of vertices within a factor of $\alpha$. While general temporal graphs may not admit sparse temporal $\alpha$-spanners for any value of $\alpha$, temporal cliques are known to admit temporal $(2k-1)$-spanners of size $\widetilde{\mathcal{O}}(kn^{1+1/k})$ for every positive integer $k$. We present a simple recursive
algorithm that computes, for every temporal clique on $n$ vertices, a temporal $3$-spanner of size $n^{1+2/\sqrt{\ln n}}=n^{1+o(1)}$, thereby improving the previous best upper bound of $\widetilde{\mathcal{O}}(n^{3/2})$. We also show that a modified version of our
algorithm computes temporal $3$-spanners of size $\mathcal{O}(nL)$ when the lifetime is bounded by $L$, i.e., all time labels are in $\{1,\ldots,L\}$, thus improving the previous bound of $\mathcal{O}(2^Ln\log n)$. Both results are particularly striking in light of the known lower bound of $\Omega(n^2)$ on the size of temporal $2$-spanners, which already holds for temporal cliques of lifetime $L\geq 3$. Both
algorithms rely on a new simple recursive decomposition that certifies temporal connectivity for a large collection of source-target pairs using only $\mathcal{O}(n)$ carefully selected edges and recursively processes only the remaining pairs. Besides yielding substantially improved upper bounds, this approach is significantly simpler than previous constructions.
Read the full abstract ▾
All authors Julia Baligacs, Davide Bilò, Václav Blažej, Maël Dumas, Anna Zych-Pawlewicz
Machine Learning
Post-Training Language Models for Gold-Medal Performance in Coding Competitions
Aleksander Ficek, Sean Narenthiran, Mehrzad Samadi et al. (+2 more)
Competitive programming has become a key test of large
language model
reasoning, with international competitions such as IOI and ICPC representing its most challenging settings. We present an end-to-end specialization pipeline combining large-scale problem curation, synthetic
reasoning traces, supervised fine-tuning (SFT), and reinforcement learning (RL). Using 22,000 curated problems, we train Nemotron-3-Nano-CC (30B-A3B) with SFT and RL and Nemotron-3-Ultra-CC (550B-A55B) with SFT alone. We further introduce GenCorrect, a feedback-driven test-time compute strategy that iteratively generates, evaluates, and refines diverse solutions. On IOI 2025, Nano-CC improves from 130 points to 291 after post-training and to 468 with GenCorrect, exceeding the gold threshold of 438.3 while Ultra-CC reaches 502. Guided by these results, we develop a competition-specific Ultra-CC system and evaluate it prospectively during IOI 2026. Under the same time, internet-access, and submission constraints as human contestants, it scores 535.4 out of 600, exceeding both the gold threshold of 361.12 and the top human score of 498.27. To our knowledge, this is the first AI system to outscore the highest-scoring human contestant on an IOI problem set.
Read the full abstract ▾
All authors Aleksander Ficek, Sean Narenthiran, Mehrzad Samadi, Somshubra Majumdar, Boris Ginsburg
Computer Vision
RoGe: Novel View Synthesis via End-to-End Implicit Reconstruction and Generation
Xiaolei Lang, Ze Kang, Zehao Huang et al. (+1 more)
Novel view synthesis from sparse inputs requires both
geometric grounding from the observed views and generative priors of unobserved regions, motivating recent hybrid methods that combine reconstruction and generation. However, existing methods bridge the two with rendered images or explicit 3D representations such as point maps or 3D Gaussians. Generation is thus conditioned on a lossy and imperfect projection of the scene, inheriting its errors, and reconstruction receives no signal from generation to correct them. We present RoGe, an end-to-end unified reconstruction and generation framework that removes this explicit bridge. It targets roaming within a scene anchored by sparse views: given a few posed images and a camera trajectory, it synthesizes a temporally coherent
video along that trajectory. From the sparse input views, RoGe builds an implicit scene representation with a feed-forward reconstruction model, and queries it with camera rays to obtain per-view
geometric features. These features are injected into a
video diffusion model as conditioning, without any explicit 3D intermediate. Both modules are trained jointly, so the generation objective directly shapes its own
geometric conditioning. We conduct extensive experiments, where RoGe outperforms reconstruction-based, generation-based, and hybrid baselines in terms of image-level quality and
video-level temporal and
geometric consistency. Ablations confirm that ray-queried implicit features outperform both raw reconstruction
tokens and rendered RGB as conditioning, and that joint training brings further gains. Our code will be released on https://jerry-locker.github.io/roge/.
Read the full abstract ▾
All authors Xiaolei Lang, Ze Kang, Zehao Huang, Naiyan Wang
Machine Learning
UE5M3 FP4 Block Scaling for Stable Language Model Pretraining
Robert Hu, Carlo Luschi and Paul Balanca
Stable 4-bit floating-point (FP4) pretraining is difficult because the E2M1 payload represents only a narrow range of magnitudes. NVIDIA's
Transformer Engine \nv{} recipe addresses this with current-tensor scaling, a randomized Hadamard transform (RHT), and bfloat16 (BF16) final layers, adding work outside the FP4
matrix multiplications. We instead pair E2M1 payloads with unsigned E5M3 (\ue{}) block scales. Their wider range permits periodic tensor scaling, while our recipe applies selective
stochastic rounding to backward gradients, omits RHT, and uses FP4 in all eligible internal linears. We pretrain a Nemotron-H 8B model for nearly 190 billion
tokens. Compared with
Transformer Engine \nv{}, the proposed block-16 recipe finishes with lower final-window training loss and, under their respective quantized-inference policies, lower validation loss measured as held-out negative log-likelihood. Its quantized-inference downstream point estimates are also higher on all three reported aggregates. A native \nv{} execution ablation that jointly removes RHT and the BF16 final-block exemption increases measured model-body
token throughput by 21.2%. These results demonstrate end-to-end software-emulated \uefp{} pretraining with a simpler recipe and motivate native support for \ue{} block scaling.
Read the full abstract ▾
Computer Vision
Efficient All-in-One Weather Restoration using Spectral Harmonization
Paula Garrido-Mellado, Daniel Feijoo, Yuning Cui et al. (+2 more)
Adverse weather conditions such as rain, haze, and snow significantly degrade image quality, posing challenges for both human perception and physical AI. Existing restoration methods require large computational budgets, struggling to process high-resolution images and handle different degradations. In this paper, we present Frequency Reconstruction via Spectral Harmonization, a novel lightweight all-in-one restoration method that explicitly decomposes feature representations into high- and low-frequency components at each scale of a hierarchical encoder-decoder architecture. By combining spectral decomposition with spatial processing through
Fourier-based skip connections, FReSH-IR captures complementary frequency information without sacrificing spatial detail. Our approach achieves similar restoration quality with 80% fewer parameters and operations than
transformer-based models. Extensive experiments demonstrate that our method offers a great efficiency-performance trade-off, highlighting its practical applications in constrained-resource systems.
Read the full abstract ▾
All authors Paula Garrido-Mellado, Daniel Feijoo, Yuning Cui, Alvaro Garcia, Marcos V. Conde
Systems and Control
Tests on the POD-P controller of INELFE Spain-France VSC-HVDC interconnector
Javier Renedo, Agustín Diaz-García, Gilles Torresan et al. (+11 more)
INELFE interconnector consists of a 2x1000 MW high voltage direct current system based on voltage source converters (VSC-HVDC) interconnecting France and Spain. INELFE VSC-HVDC link is embedded into the high voltage alternating current (HVAC) system. Electromechanical oscillations, also known as power oscillations, are a major concern worldwide. INELFE VSC-HVDC link has specific controllers to damp power oscillations by modulating active (P)- and reactive (Q)-power injections of the VSC converters (POD-P and POD-Q controllers, respectively). The Spanish and French Transmission System Operators (TSOs) carried out a join task force to - Increase the gain of POD-P controller of INELFE VSC-HVDC link. - Make it possible to use the POD-P controller together with angle difference control (ADC) of INELFE VSC-HVDC interconnector. The objective of these modifications is to improve the effectiveness of POD-P controller and, therefore, to increase its contribution to the damping of inter-area oscillations in the Continental Europe (CE) power system. Such changes require (a) extensive simulation studies and (b) extensive tests in different operation modes, in order to ensure the correct behavior of the system. This paper presents simulation studies and field tests on the POD-P controller of INELFE VSC-HVDC interconnector in different modes of operation.
Read the full abstract ▾
All authors Javier Renedo, Agustín Diaz-García, Gilles Torresan, Eduardo Lorenzo Cabrera, Antonio Cordón, Silvia Sanz Verdugo, Juan Peiró, Javier Pérez Castro, Gorka Calvo, Aitor Hernández Sautua, Ambroise Petit, Hamza Oukhayi, Samy Akkari, Hani Saad
Sound
Understanding Automatic Mixing: A Subtask-Oriented Analysis of Two-Stage Mixing System
Jinjie Shi, Wei Hua, Kunzhu Xie et al. (+3 more)
Automatic mixing transforms multitrack recordings into perceptually coherent, balanced, and aesthetically consistent mixes. In real-world production, this task is challenging due to large track counts, diverse instrumentation, and strong inter-track dependencies. Two-stage systems address this complexity by separating intra-group processing from inter-group mixing, yet it remains unclear whether their gains arise from stronger component models or from explicit task decomposition. We present a subtask-oriented analysis of automatic mixing through three controlled listening experiments. We investigate whether full-mix models transfer to intra-group mixing, whether downstream models compensate for grouping and loudness errors, and whether two-stage decomposition improves full-mix quality. Across three dense pop and rock excerpts, transfer differs between the evaluated models; inappropriate grouping causes clear downstream degradation, while altered loudness relationships have weaker and model-dependent effects. Both two-stage variants significantly outperform their corresponding single-stage baselines. These findings support explicit separation of local balance and global mix coordination as a useful design principle for automatic mixing. Code and audio examples are available online.
Read the full abstract ▾
All authors Jinjie Shi, Wei Hua, Kunzhu Xie, Make Li, Yuchen Liu, Joshua Reiss
Machine Learning · Cross-listed from physics.comp-ph
Learning Spectrally Optimised Mesh-Free Discretisations
Lucas Gerken Starepravo, Henry Broadley, Steven Lind et al. (+1 more)
Numerical methods for partial differential equations (PDEs) discretise differential operators, ideally reproducing the action of the continuous operators across all wavenumbers permitted by a given discretisation. Spectral-type methods approach this ideal, but rely on structured grids or high-order meshes that are hard to generate for complex geometries. In contrast, mesh-free methods are geometrically flexible, yet how faithfully they reproduce the operator across resolved scales is strongly influenced by a heuristically chosen kernel, which selects one of many weight sets satisfying the same consistency conditions without regard to the resulting spectral response. To address this, we introduce Spectrally optimised Neural Discretisations (SpeND), a framework that learns the map from local stencil geometry to discretisation weights on unstructured point clouds. A projection layer constrains every predicted stencil to the affine set defined by the discrete moment conditions, so
polynomial consistency, and hence formal order of accuracy, holds exactly. Since consistency is enforced by the architecture, the weights can additionally be optimised for accuracy on a prescribed function space. Herein, we use
Fourier modes and target the exact differentiation response over a chosen wavenumber band, so training is unsupervised. The loss function is a design interface: changing how it weights wavenumbers yields operators with distinct accuracy profiles. The trained discrete differential operators are PDE-agnostic and are applied without retraining to the Poisson, Burgers and Navier–Stokes equations, to near-boundary stencils, and across resolutions. At the same order and resolution, SpeND matches or improves on established mesh-free discretisations, and has been shown to reduce the wall-clock time between $3-20\times$ to achieve equivalent error as the baselines.
Read the full abstract ▾
All authors Lucas Gerken Starepravo, Henry Broadley, Steven Lind, Jack R. C. King
Computer Vision
Benchmarking RAW and RGB Restoration in Image Signal Processors
Zihao Lu, Radu Timofte and Marcos V. Conde
Modern cameras transform RAW sensor measurements into sRGB images through an image signal processor (ISP). We
benchmark two placements for blind restoration around a fixed ISP: (A) pre-ISP restoration in the RAW domain and (B) post-ISP restoration in the sRGB domain. The
benchmark covers four smartphone device groups, two learned ISPs, three degradation regimes–noise, blur, and joint noise and blur–, and several representative RAW and RGB restoration models. Our results show that placement alone does not determine performance. The RAW restoration strategy outperforms the best generic RGB restoration models. However, RGB restoration models trained considering the ISP transformations, achieve the best overall performance. Our novel
benchmark demonstrates that the image reconstruction performance strongly depends on the
alignment between the restoration model and the target imaging pipeline. We consequently recommend reporting restoration placement and ISP-aware supervision as key experimental factors. Our code is available at https://github.com/mv-lab/AISP
Read the full abstract ▾
Robotics
Toward Robust LiDAR Semantic Segmentation for Real-World Deployment: Evaluation under Coarse Labels, Adverse Conditions, and Domain Shifts
Samir Abou Haidar, Alexandre Chariot, Mehdi Darouich et al. (+2 more)
LiDAR-based semantic
segmentation is a core perception module for autonomous vehicles and mobile
robots. Despite the strong performance of recent state-of-the-art methods on standard
benchmarks, existing evaluation protocols remain focused on clean, single-domain settings and fine-grained label taxonomies, leaving deployment readiness largely unassessed. Real-world systems must handle safety-critical label semantics, degraded sensing conditions, and cross-domain variability, yet no unified protocol currently addresses all three aspects together. In this paper, we propose a structured evaluation protocol that assesses the deployment readiness of
LiDAR semantic
segmentation models along three complementary dimensions: (i) coarse-label evaluation aligned with autonomous driving safety priorities, revealing how label granularity affects different methods; (ii) robustness under eight types of
LiDAR corruptions designed to emulate real-world atmospheric,
geometric, and sensor degradations; and (iii) domain generalization across
datasets without adaptation. The evaluation includes inference speed measured on an embedded Jetson AGX Orin platform, directly reflecting deployment constraints. Our results show that fine-grained
benchmark rankings do not always reflect safety-relevant performance, that all methods experience substantial degradation under corruptions with architecture-dependent robustness characteristics, and that current domain generalization remains insufficient for reliable deployment. These findings expose concrete gaps between
benchmark performance and deployment readiness, and provide a reference protocol for more practically grounded evaluation of
LiDAR semantic
segmentation.
Read the full abstract ▾
All authors Samir Abou Haidar, Alexandre Chariot, Mehdi Darouich, Cyril Joly, Jean-Emmanuel Deschaud
Artificial Intelligence
AI Contextual Measurement for Recovering Individual and Group-Level Effects: Validation Against Survey Measures and an Occupational Application
Wenxin Jiang, Xuyang Wang and Yuxiao Wu
Researchers increasingly use artificial intelligence to construct measures of social, organizational, and occupational characteristics that are absent from conventional surveys. We propose AICOME, AI COntextual MEasurement, a framework for evaluating whether AI-derived respondent-level measures can recover individual and group-level effects in contextual models. The key idea is that an AI measure constructed at the respondent level can be used to derive its group-level aggregate and its individual deviation, allowing researchers to estimate both between-group and within-group associations rather than treating AI measurement as response prediction alone. We validate the framework using the 2022 China Family Panel Studies (CFPS), where occupations provide the empirical grouping structure and several job-related survey variables provide validation
benchmarks. For computer use, foreign-
language use, weekly hours, and management responsibilities, we compare survey measures with AI-derived measures in response-level, model-level, contextual, and boundary-condition validations. The results show that AI contextual measurement can recover much of the contextual-model information contained in observed survey variables when rich respondent and job characteristics are available. Weekly hours provides the strongest validation case, with AI-derived measures reproducing the large negative between- and within-occupation associations with satisfaction observed in CFPS. The framework also identifies clear boundary conditions: performance deteriorates when information is restricted to occupation and basic demographics, and recovery is weaker when several related concepts are treated as simultaneously unobserved. The findings suggest that AICOME is most useful for recovering a limited number of theoretically important constructs from rich existing
datasets.
Read the full abstract ▾
Machine Learning
Cliff: Learning Process Rewards from the First Mistake
Peixuan Han, Runhui Wang, Ketan Ramaneti et al. (+3 more)
Reinforcement learning with verifiable rewards (RLVR) has emerged as a powerful paradigm for large
language model (LLM) post-training, but its reliance on coarse outcome rewards leads to limited guidance on intermediate
reasoning processes. Existing approaches such as process reward modeling and on-policy
distillation introduce additional constraints, such as reliance on a specialized reward model or assuming identical
reasoning patterns between teacher and student. Nevertheless, we observe that once a
reasoning process first goes wrong, evaluating the subsequent
reasoning provides limited additional information, as it is already conditioned on an invalid prefix. Therefore, we propose Cliff, a reward shaping strategy that utilizes an off-the-shelf LLM as a teacher to identify the first mistake in each rollout. As a result, the rollout is naturally decomposed into two parts: a correct prefix and an incorrect suffix. Cliff then converts this signal into
token-level advantages, assigning positive advantages for the correct prefix and negative feedback afterward. Experiments across 12 different scenarios demonstrate that Cliff consistently improves
reasoning performance, outperforming on-policy
distillation by 15% and standard GRPO by 7%, even with teachers of modest capability. Furthermore, we analyse the role of “ground truth” in Cliff and investigate its training dynamics. These results establish Cliff as a simple, general and effective approach for improving RLVR with richer, fine-grained supervision.
Read the full abstract ▾
All authors Peixuan Han, Runhui Wang, Ketan Ramaneti, Jie Hao, Gerald Friedland, Chris Kong
Computer Vision
GDB-Reward: From Evaluation Metrics to Training Rewards for Graphic Design
Adrienne Deganutti, Purvanshi Mehta, Simon Hadfield et al. (+1 more)
Text-to-image models excel at natural image synthesis but struggle with graphic design, where success depends on satisfying precise constraints on typography, layout, color, and visual communication. While
prompt optimization offers an attractive alternative to expensive
diffusion model fine-tuning, learning
prompts for frozen image generators requires informative reward functions despite the entirely non-differentiable generation process. Reinforcement learning does not require differentiable objectives; it requires only scalar rewards capable of ranking candidate outputs. This raises a simple question: can design evaluation metrics themselves become reinforcement learning rewards? Our central contribution is GDB-Reward, a framework that systematically transforms heterogeneous graphic design evaluation metrics into a unified reinforcement learning reward. Experiments demonstrate that GDB-Reward provides an effective optimization objective, substantially improving adherence to the design specification in perceptual quality, rendering fidelity, and spatial accuracy while keeping the image generator entirely frozen. More broadly, our results demonstrate that heterogeneous, non-differentiable evaluation metrics can move beyond passive benchmarking to become effective optimization objectives for reinforcement learning in domains where differentiable supervision is unavailable.
Read the full abstract ▾
All authors Adrienne Deganutti, Purvanshi Mehta, Simon Hadfield, Andrew Gilbert
Robotics
Do Better Imagined Rollouts Mean Better Robot Control? A Controlled Study of World-Model Evaluation Under Feedback
Dharini Raghavan and Amritpal Singh
Predictive models are increasingly used in robotics for state estimation, planning, control, and policy evaluation, yet they are often judged by open-loop prediction accuracy over a fixed horizon. In closed-loop operation, a
robot repeatedly acts, receives new measurements, updates its state estimate, and recomputes control. We study this difference in a differential-drive path-tracking task with biased odometry and intermittent landmark sensing. Six state
estimators are evaluated across 24 sensing conditions using trajectory replay, a 20-step measurement-free rollout, and closed-loop tracking. Replay position RMSE correlates more strongly with closed-loop cross-track RMSE than rollout error (Spearman rho = 0.923 vs. 0.774) and selects a different
estimator from the closed-loop optimum in 5/24 conditions, compared with 18/24 for the rollout metric. We then vary rollout horizon and measurement-update interval. With H=20, rank agreement decreases from rho = 0.916 with measurements at every step to rho = 0.774 with no measurements. A horizon-update grid shows that long prediction horizons remain informative when regular corrections are retained, whereas long rollouts without correction can produce rankings that differ substantially from closed-loop behavior. We also test recurrent
estimators trained on longer sensing outages. This improves the EKF-anchored models under combined sensing degradation, reducing GRU-EKF cross-track RMSE from 1.72 m to 1.06 m, but the gain is not consistent across isolated outages or
estimator architectures. These results show that predictive-model evaluation in robotics should specify both prediction horizon and measurement-update schedule. For models used in feedback, offline rollouts are most informative when their sensing and correction pattern reflects closed-loop operation. Code is available at https://github.com/rdharini2001/
Robot_World_Model
Read the full abstract ▾
Systems and Control
Guidelines for the implementation of power oscillation damping controllers in power converters
Javier Renedo, Macarena Martín Almenta, Sergio Martínez Villanueva et al. (+3 more)
One of the most effective ways to damp electromechanical oscillations in power systems is by means of supplementary controllers attached to the different devices in the power system is by means of power system stabilizers (PSS) in synchronous machines or by means of power oscillation damping (POD) controllers in facilities with power converters. In the recent years, the use of POD controllers in voltage source converters (VSCs) with grid-following (GFL) control has been investigated. Although the potential of POD controllers to help to damp inter-area oscillations in power systems is enormous, their correct implementation is not trivial, because their effectiveness is strongly linked to their settings. This paper provides guidelines for the implementation of POD controllers in power converters for application in real-world power systems. The paper proposes compliance criteria for POD controllers using a synthetic test system and a systematic methodology used in Spanish technical standard for monitoring compliance (NTS), considering practical considerations. The paper also includes numerical examples to illustrate compliance criteria for POD controllers in a synthetic test system. A generic power converter with grid-following (GFL) control is used for the analysis by simulation and POD controllers using modulation of active-power injection (POD-P), reactive-power injection (POD-Q) or both simultaneously (POD-PQ) will be analysed. Results were validated in a large-scale power system. The paper concludes that by using appropriate synthetic systems, methodologies and compliance criteria, POD controllers in power converters could be effective to damp electromechanical oscillation in large-scale power systems.
Read the full abstract ▾
All authors Javier Renedo, Macarena Martín Almenta, Sergio Martínez Villanueva, Agustín Díaz-García, Antonio Cordón, Davide Gotti
Artificial Intelligence
Large Language Models (LLMs) for Telecom Root Cause Analysis (RCA): A Structured Reasoning Framework for Evidence-Grounded Diagnosis
Hao Zhou, Mandar Kulkarni, Hao Chen et al. (+2 more)
Root cause analysis (RCA) is a critical task in telecom network operations, but diagnosing performance degradations in modern 5G and emerging 6G networks remains challenging due to complex cross-layer dependencies. While large
language models (LLMs) offer promising capabilities for
reasoning and knowledge integration, directly applying vanilla LLMs to telecom RCA often leads to
hallucination, unstable
reasoning, and poor
alignment with structured network evidence. This work first reviews the
evolution of telecom RCA from rule-based and machine learning (ML) approaches to emerging LLM-enabled techniques, and provides an overview of recent paradigms, including structured
reasoning,
retrieval-augmented knowledge grounding, agentic orchestration, and verifiable
reasoning. Building upon these insights, we propose a structured
reasoning framework for LLM-enabled telecom RCA that aligns diagnostic
reasoning with telecom-specific evidence and domain knowledge. The proposed approach first organizes heterogeneous network telemetry into canonical contexts, and then enforces decision-path
reasoning during diagnosis, and finally generates evidence-grounded explanations for reliable fault identification. Experimental results on two 5G RCA
datasets, TeleLogs and TelecomTS, demonstrate that the proposed framework consistently improves diagnostic accuracy and decision consistency compared with baseline techniques. These cross-
dataset results highlight the importance of structured
reasoning design for practical LLM-based RCA systems in next-generation telecom networks.
Read the full abstract ▾
All authors Hao Zhou, Mandar Kulkarni, Hao Chen, Yan Xin, Charlie Zhang
Discrete Mathematics
frb100-40 After Two Decades: An Optimality Certificate and a Preregistered Search Study
Onur Uğurlu
For more than 20 years, the Model-RB
benchmark frb100-40 remained an open challenge; since 2014, its public record had stood at 99 of 100 variables. We give a directly checkable 100-vertex independent set for its 4,000-vertex graph. Together with a verified partition into 100 cliques of size 40, the witness proves that the maximum independent-set size is 100 and the minimum vertex-cover size is 3,900. The
stochastic run that found the witness is kept separate from this proof. We evaluated its added pair and triple repair operators in a preregistered campaign comprising 8,668 valid runs. The primary comparison found no detectable acceleration over base ULSA (hazard ratio 0.967, 95% confidence interval 0.915-1.023; p=0.248), and the factorial ablation reached the same conclusion. On a smaller FRB suite, the group-aware CSP pipeline solved 2,500/2,500 runs, compared with 2,391/2,500 for LibMVC-NuMVC. On frb100-40, full ULSA, base ULSA, and NuMVC each produced 0/56 new certificates. With no events, the planned cross-solver hazard ratios remain unidentified. NuMVC ended with cover size 3,902 in 40 runs and 3,903 in 16. Exhaustive enumeration showed that none of the 108 unique recorded conflict-two states had a strictly improving group-aware CSP neighbor within Hamming radius three. The certificate settles the instance. The experiments characterize the search barrier, and the preregistered comparisons show no heuristic advantage.
Read the full abstract ▾
Computer Vision
AutoCompass: Accurate Visual Localization on Public Maps by Learning from Weak Labels
Javier Tirado-Garín, Alan Savio Paul, Shuai Chen et al. (+5 more)
Neural map matchers estimate an image's 3-DoF pose relative to a 2D map. These models are trained on large-scale
datasets of geo-referenced images, whose position and heading labels often contain noise that affects the trained models. To address this, we present AutoCompass, a supervision approach for training neural map matchers from inaccurate absolute pose labels. First, we show that heading labels are unnecessary: trained from raw GPS labels, models learn to predict accurate headings, automatically. Second, defining a tolerance region around raw GPS improves positional accuracy. Third, if available, our supervision uses relative poses between training images, obtained via SLAM or SfM, which provide a more accurate training signal. Across driving and
egocentric benchmarks, AutoCompass consistently outperforms counterparts trained with the usual strong reliance on absolute pose labels.
Read the full abstract ▾
All authors Javier Tirado-Garín, Alan Savio Paul, Shuai Chen, Axel Barroso-Laguna, Tommaso Cavallari, Daniyar Turmukhambetov, Victor Adrian Prisacariu, Eric Brachmann
Artificial Intelligence · Cross-listed from econ.GN
Dutch Books for Language Models
Isaiah Andrews and Suproteem Sarkar
People increasingly use
language models to support life decisions. Many such decisions involve a probabilistic forecast: How likely is a major life event, a natural disaster, or an economic outcome? Users of
language models may implicitly trust that these forecasts fall out of a coherent world model. In this paper, we evaluate the coherence of
language model probabilistic forecasts through a procedure that builds on a
theorem due to de Finetti. We elicit forecasts from
language models across events generated from stock returns data. We then use linear programs to compute the largest Dutch-book profit - the profit an arbitrageur could guarantee by betting against model-generated probabilities - which we use as a measure of incoherence. Our procedure does not require outcome labels, so we can evaluate coherence even in settings where outcomes are not observed or have not yet resolved. We find substantial evidence of incoherence in
language model forecasts. Such incoherence increases when there are richer logical relationships between events, and irrelevant contextual details can increase incoherence by an order of magnitude. We conclude by discussing how alternative training strategies may improve probabilistic coherence.
Read the full abstract ▾
Computation and Language
DiscoSign: Discourse-Aware Text to Sign Language Gloss Translation
Vasileios Baltatzis, Mert Inan, Connor Gillis et al. (+4 more)
Sign
language processing systems have traditionally operated at the sentence level, ignoring critical discourse phenomena fundamental to sign
language comprehension. We introduce DiscoSign, a computational approach for discourse-aware text to sign
language gloss translation grounded in linguistic research. We address three key phenomena within our modular Large
Language Model (LLM)-based translation framework: (i) spatial coreference resolution, where entities maintain consistent spatial locations throughout discourse; (ii) Question-Answer Clauses (QACs), pseudocleft structures serving specific discourse functions; and (iii) concept-gloss consistency, ensuring stable mappings between English concepts and American Sign
Language (ASL) signs. Traditional translation metrics fail to capture discourse-level quality, so we introduce a suite of novel evaluation metrics designed to assess each dimension of discourse coherence addressed by our framework. Experiments on sentence-level and discourse-level
datasets show that our approach for discourse-aware processing significantly improves spatial consistency and entity tracking relative to sentence-only translation, while maintaining competitive single-sentence gloss translation quality. Our work establishes the first systematic framework for discourse-level text to sign
language gloss translation with corresponding evaluation methodology.
Read the full abstract ▾
All authors Vasileios Baltatzis, Mert Inan, Connor Gillis, Raja Kushalnagar, Lorna Quandt, Leah Findlater, Colin Lea
Machine Learning · Cross-listed from stat.ML
Full-Model Optimality for Tunable Linear Generative Priors in Compressed Sensing
Zhaoming Li and Paul Hand
Generative models have been studied experimentally and theoretically as priors for inverse problems such as compressed sensing. Recent work by Gunn et al. studied the use of generative priors with tunable complexity, where a family of generative priors with varying complexity is maintained and a specific complexity can be selected at inversion time. They demonstrated that lower reconstruction errors can be experimentally attained for a variety of inverse problems by appropriately tuning the complexity of the generative prior. In the present paper, we establish theory for compressed sensing in the setting of a tunable family of linear generative priors naturally related through their singular value decompositions. We prove that in noiseless Gaussian compressed sensing, the full-dimensional linear prior attains the minimum expected reconstruction error over the entire family of linear priors. Thus, in this idealized linear noiseless setting, tuning to a lower-complexity prior does not improve the expected reconstruction error. This result is in contract to the behavior of denoising, where lower complexity priors attain lower reconstruction errors due to a standard bias-variance tradeoff. This result indicates that the experimental benefits of tunability in compressed sensing with neural network priors arises due to nonlinearities in the generative models.
Read the full abstract ▾
Software Engineering
ShikumiMiner: Mining Recurring Implementation Patterns in AI Codebases
Afsana Tasnim, Saif Uddin Mahmud, Md Erfan et al. (+1 more)
Large
language models are paving the way towards innovation by understanding, analyzing, summarizing and generating content in the modern world. Currently there are thousands of LLM projects developed by engineers in open-source repositories. However, whether these LLM projects have underlying patterns or not remains a question. Exploring these underlying patterns will give new dimensions to the developers who aim to develop these LLM projects. In this paper, we propose ShikumiMiner, a static-analysis framework that combines Abstract Syntax Tree (AST) and Control Flow Graph (CFG) features to detect and compare recurring implementation patterns in C++ local LLM codebases. We analyze ten GitHub open-source repositories and classify functions into seven study-specific categories using a multi-label Random Forest model. Studying these patterns can provide useful insights for developers aiming to design LLM applications.
Read the full abstract ▾
All authors Afsana Tasnim, Saif Uddin Mahmud, Md Erfan, Md Rayhanur Rahman
Artificial Intelligence
SafeEvolve: Harness-Policy Co-Evolution from Agent Experience for Safety Alignment
Qinghua Mao, Wanying Qu, Dadi Guo et al. (+8 more)
The performance of LLM-based
agents is jointly shaped by the base model and the harness used when interacting with the environment. This exposes them to safety risks in both harmful final responses and multi-step execution trajectories. Existing safety
alignment mechanisms often rely on either external harness updates or policy optimization, yet applying either paradigm in isolation fails to bridge runtime control with intrinsic safety. We propose SafeEvolve, an experience-driven self-evolving framework for
agent safety
alignment. SafeEvolve leverages safety experience from completed on-policy trajectories to drive a continual loop of harness-policy co-
evolution. On the harness side, SafeEvolve converts trajectory-level safety evidence into bounded, component-level updates across safety
prompt and hierarchical skills, yielding auditable and reversible harness artifacts. On the policy side, SafeEvolve follows a two-stage SFT-RL paradigm, where harness-use SFT bootstraps the policy to actively leverage evolved harness artifacts, and harness-augmented RL further shapes autonomous safety behaviors during multi-step exploration via verifier-decomposed rewards. Through harness-policy co-
evolution, SafeEvolve converts safety experience into an evolved runtime harness and improved policy behavior. Experiments on agentic safety
benchmarks show that SafeEvolve achieves a stronger safety-utility tradeoff than existing baselines. For Qwen3.5-4B, SafeEvolve achieves a $3\times$ ASR reduction on AgentDojo while improving benign utility from 59.79% to 61.86%.
Read the full abstract ▾
All authors Qinghua Mao, Wanying Qu, Dadi Guo, Leitao Yuan, Qingyu Liu, Yu Li, Guanxu Chen, Yanwei Fu, Xi Lin, Xia Hu, Dongrui Liu
Computation and Language
EarlyEval: Cheaper Agent Evaluation via Early Outcome Prediction
Yuling Shi, Zhensu Sun, Junsen Dong et al. (+3 more)
Evaluating LLM
agents is essential for guiding their development, yet it has grown prohibitively expensive: a single pass of a frontier model over an agentic
benchmark can cost hundreds to thousands of dollars, a price paid repeatedly across iterative development cycles. Prior efforts, centered on
benchmark distillation, reduce the number of evaluation tasks but leave the cost of executing each retained task untouched. In this work, we introduce early outcome prediction, a complementary axis of efficiency that instead cuts cost within each task. Our key insight is that an
agent's final outcome is often evident from its intermediate behavior well before execution completes. We instantiate this idea in EarlyEval, a lightweight framework that trains a pair of LightGBM success and failure classifiers over behavioral, textual, and reference-solution features, and halts an
agent run the moment either classifier crosses a calibrated confidence threshold, adding negligible per-step overhead. Across three
benchmarks, SWE-bench Verified, TerminalBench, and Toolathlon, EarlyEval can eliminate 13%-26% of
agent steps and up to 44.1% input
tokens and 29.4% output
tokens at 89%-97% prediction accuracy, while perturbing per-
agent resolve rates by only one to two percentage points on average.
Read the full abstract ▾
All authors Yuling Shi, Zhensu Sun, Junsen Dong, Chengcheng Wan, David Lo, Xiaodong Gu
Software Engineering
Type Hints in Python Libraries and Frameworks: An Empirical Analysis of Adoption and Maintenance
Thiago Roberto Magalhães, Fabio Petrillo and João Eduardo Montandon
Context: In Python, type hints allow developers to annotate variables and functions with explicit type information, improving code clarity and reliability. Although type hints are widely available, little is known about how they are adopted and maintained in libraries and frameworks. Objective: We investigate the adoption, usage, maintenance, and rationale of type hints in Python libraries and frameworks. Method: We analyzed 1,000 popular GitHub repositories, identifying libraries and frameworks and extracting their type annotations. We examined annotation coverage, the locations and origins of annotations, their
evolution across git histories, and the relationship between developer annotations and types inferred by Pyright. Results: Of the analyzed repositories, 91% of libraries use type hints at least once, although adoption is inconsistent. Among libraries with systematic usage, maintainers prioritize function parameters and return types, with median coverage of 45.8% and 35.9%, respectively, and mainly use built-in types (73.0%). When modified, annotations tend to migrate to more expressive types. Developers annotate members even when Pyright can infer their types, and these annotations often simplify the inferred type. Conclusion: Type hints in Python libraries and frameworks primarily serve as API contracts rather than comprehensive descriptions of implementation details. The findings suggest opportunities for tooling that prioritizes public interfaces, identifies meaningful annotation changes, and supports maintainers in evolving type information.
Read the full abstract ▾
Computer Vision
ShallowStream: Index Shallow then Answer Deep for Streaming Video Understanding
Jitai Hao, Ke Yang, Di Yan et al. (+3 more)
Streaming
video understanding is a critical capability for real-world applications, including
embodied intelligence, autonomous driving, industrial monitoring, surveillance and early warning, and wearable assistants. However, processing continuous
video streams with
multimodal large
language models (MLLMs) is computationally expensive. Existing efforts have explored reducing streaming overhead through visual
token pruning,
token merging, quantization, on-demand frame
retrieval, and context offloading. However, most existing methods overlook the dimension of model depth. Repeatedly executing full-depth MLLM prefill over incoming frames is prohibitively expensive, incurring substantial computational overhead and causing the KV cache to grow at a rate directly proportional to the prefill depth. To address these challenges, we propose ShallowStream, a novel framework that leverages the shallow layers of an MLLM to simultaneously perform frame encoding and
retrieval index building. During stream processing, ShallowStream maintains an always-on lightweight index using the KV cache of shallow layers. During query-time answering, we leverage the
attention scores generated by the shallow layers to score context frames and employ a diversity-aware selection strategy to retrieve precise and comprehensive evidence. ShallowStream achieves performance on par with the strongest existing streaming methods, while reducing per-frame prefill latency and 10-second end-to-end latency by up to 52.1x and 11.9x, respectively. Our code is available at https://github.com/CURRENTF/ShallowStream.
Read the full abstract ▾
All authors Jitai Hao, Ke Yang, Di Yan, Fan Liu, Qiang Huang, Jun Yu