In graph-level learning, graph neural networks are typically applied with a fixed computational template: the same depth, local message-passing operator, and global attention mechanism are used for every input graph. However, different graphs within the same dataset can vary substantially in prediction difficulty and may require different mixtures of local and non-local computation. We study this phenomenon as graph-level heterogeneity in both classification and regression tasks. To address it, we propose SiGMA (Sigmoid Gated Message-Passing and Attention), a gated hybrid graph neural network architecture that combines message-passing and attention heads within each layer. Inspired by the growing use of gated attention and gated computation in large language models and vision transformers, each head in SiGMA is modulated by a learned node-wise sigmoid gate, allowing the model to adaptively regulate which local or global mechanisms are active for different nodes, layers, and input graphs. The resulting architecture can emulate expressive message-passing backbones, admits near-identity computation paths that are not forced to over-smooth, and enables direct non-local communication beyond message-passing neighborhoods to mitigate over-squashing. Empirically, SiGMA achieves strong performance across graph-level benchmarks, consistently matching or improving over competitive graph transformer and hybrid baselines. Our results suggest that adapting computation to graph-level heterogeneity is a promising direction for building more flexible and effective graph learning architectures.
$π^4$ is a unified foundation model for 4D scene perception. From diverse, uncalibrated RGB observations in the wild---a single image, an unordered collection, or a video sequence---it recovers dense geometry, camera poses, and metric scale, together with 2D/3D point tracks and moving-object masks for multi-image inputs. It couples a geometry branch with a geometry-conditioned motion branch, which predicts image-space correspondences that are lifted into 3D using reconstructed target-view geometry. Across diverse benchmarks, $π^4$ achieves the best overall average rank in geometry evaluation, alongside competitive performance in single- and multiview 2D/3D point tracking and moving-object segmentation. With a shared backbone, the full geometry-and-motion system contains approximately 0.5B parameters---less than half the size of most baselines. It produces joint reconstructions and tracks in one forward pass, including large-scale scenes with thousands of input images despite training with contexts of at most 64 views. We further present the data and training recipe underlying these capabilities, together with empirical studies of key design choices. Code and models are available at https://github.com/yyfz/Pi4.
Robot manipulation videos provide a rich visual experience, yet recovering robot action and gripper-state annotations from RGB observations still remains challenging. Using the embodiment-agnostic tool-center-point (TCP) pose as the action space, we present TCP4D, a unified model in which a shared 4D scene representation both tracks observed TCP motion and generates future TCP trajectories. Given a monocular video and initial 2D TCP queries, the tracker directly recovers metric TCP poses in SE(3) and gripper states alongside scene geometry and camera parameters. The generator samples local features from the same 4D representation along the recovered TCP tracks, then combines them with robot observations and language instructions to predict future TCP states. Our new cross-embodiment augmentation pipeline converts existing RGB videos into demonstrations for alternative embodiments by retargeting the recovered motion and synthesizing the corresponding observations, without recorded source-robot states. Experiments show that the tracker achieves absolute trajectory errors of 7.60 and 16.07 mm on DROID and RoboLab and adapts to a new embodiment with only ten demonstrations. The generator achieves 69.19% mean success on RoboTwin and outperforms well-established policy models on two real-world tasks, while joint training on source and augmented demonstrations improves target-robot success by 41.5% over augmented-only training.
The "AI revolution" has enabled teams across the technology industry to deliver software features at unprecedented velocity. While code evolves faster than ever, customer-facing documentation often struggles to stay up to speed. As users increasingly rely on LLM-based agents that consume documentation as their primary interface, the impact of this documentation gap worsens: stale information propagates and is treated as authoritative. Signals for identifying where documentation has fallen behind already exist in abundance, both explicit (e.g., resolved support tickets revealing what documentation lacks) and implicit (e.g., code-documentation discrepancies detectable from automated loops). We deploy a multi-agent system, built around a triage agent we call "TriAgent," that continuously extracts these signals and repairs documentation drift. Over 43 days of shadow-mode operation, our system triaged 499 tickets and surfaced 92 gaps in the documentation. Using a panel of LLM judges from independent model families, 81.1% of TriAgent's verifiable assertions were correct (245 of the 302 assertions the human resolution addresses; majority of three judges); on length-comparable pairs the panel preferred TriAgent's response in 70.5% of pairs and the human engineer's in 23.2% (the rest had no judge majority). Importantly, agent- and human-sourced replies tended to be complementary, with the human often surfacing insights that were either buried within the existing documentation or missing from it entirely; our system turns each such insight into a documentation gap and, after human review, a fix. Repairing a gap measurably improved the agent's later answer: when the agent re-answered the same ticket after the repair, blinded judges preferred the post-repair answer over its own pre-repair answer 71.4% of the time (n=35). A blinded human audit confirms strong alignment on self-improvement.
In the first version of this paper I reported that conservative Reddit users began writing more alike after ChatGPT was released while progressive users did not. That result came from a similarity measure that rises and falls with comment length. I redo every analysis of the original paper on all 6,056,593 comments with a sentence encoder and a dispersion measure computed within length strata, and the conservative break goes away. The two camps show no reliable difference in how within-group similarity changed after late 2022, and the ChatGPT date behaves like any of the more than two thousand other cutoffs I test. The matched flair comparison, the cumulative AI index, the stayer panel and the entropy signature each lose their support once length is held fixed. The one change that survives is older. Left and right comments drew closer to each other between 2019 and 2022, and that convergence was complete before ChatGPT arrived. I end with what this implies for studies that measure homogenization with embeddings.
Visual foundation models are central to physical intelligence, yet their development remains constrained by fragmented training datasets, inconsistent annotation quality, and limited multimodal supervision. To address these challenges, we introduce GameRealm, a large-scale, richly annotated dataset that grounds visual observations in synchronized, game-native supervision. GameRealm encompasses over 120 million RGB frames and 1,000 hours of gameplay footage across all views. Beyond its unprecedented scale, GameRealm provides rich multimodal supervision, including metric depth, camera parameters, synchronized audio, control inputs, and 3D point trajectories. Its diversity further extends to flexible view configurations, spanning single-view, stereo, multi-view, and panoramic recordings. To enable scalable acquisition across diverse game environments, we develop a unified data collection pipeline that combines human-operated interaction with agentic exploration. Extensive experiments across diverse visual learning tasks demonstrate that GameRealm consistently strengthens visual foundation models in both visual generation and geometric learning. These results establish GameRealm as a versatile and valuable data resource, unlocking new opportunities for advancing visual intelligence.
Diffusion models generate realistic visual content, yet often fail to produce rare but plausible compositions. When prompted with combinations that are valid but underrepresented in training data (such as a snowy beach or a rainbow at night), the generation process frequently collapses toward more common alternatives. We identify this failure mode as default completion bias, where denoising trajectories are implicitly attracted toward high-frequency semantic configurations. Existing guidance mechanisms do not explicitly model this competing tendency and therefore struggle to prevent such collapse. We introduce Default Completion Repulsion (DCR), a training-free framework that explicitly models and suppresses default completion behavior. DCR constructs a counterfactual attractor by relaxing the rare compositional factor while preserving surrounding semantics, inducing an alternative denoising trajectory reflecting the model's preferred completion. We define the discrepancy between target and attractor trajectories as a counterfactual drift, and propose a projection-based repulsion mechanism that removes guidance components aligned with this drift direction. This suppresses undesired frequent completions while preserving other semantic components. DCR operates entirely within the standard diffusion sampling process without retraining or architectural modification. Experiments on rare compositional prompts show that DCR improves compositional fidelity while maintaining visual quality. Our analysis further shows that generations from the attractor prompt alone closely match the baseline outputs, confirming that it captures the model's default completion.
Non-invasive brain-to-text interfaces offer alternative communication pathways for people with communication disabilities. However, sentence-level decoding from electroencephalography (EEG) remains challenging. We propose a hierarchical EEG-text alignment framework that retrieves previously unseen sentences by progressively decoding coarse-grained semantic information and fine-grained sentence representations. We hypothesize that broader semantic context can be more reliably captured from EEG than individual sentences. Accordingly, EEG representations are first classified into semantic classes, which are defined as groups of related sentences, followed by sentence retrieval within the predicted class. Among over 2,000 unique text samples, our approach achieves, on unseen sentences, average top-5 and top-10 retrieval accuracies of 62.6\% $\pm$ 1.0\% and 83.5\% $\pm$ 1.0\%, respectively, compared with 0.3\% $\pm$ 0.3\% and 0.5\% $\pm$ 0.2\% for direct EEG-sentence alignment without semantic classification. Qualitative analysis further reveals that decoding performance depends strongly on textual context. These findings demonstrate the potential of hierarchical semantic alignment for advancing sentence-level EEG-to-text decoding.
Enabling large language models (LLMs) to autonomously conduct training and improve target models is an important foundation for recursive self-improvement (RSI). As LLMs expand their role from being trained to conducting training, their model-training capability warrants dedicated study and systematic development. To this end, we construct RSI-Prime, a dataset for improving model-training capability, comprising 350 complete experimental trajectories across seven task categories. To help LLMs learn to make high-quality experimental decisions from training practice, we perform fine-grained decision identification and quality annotation, then apply decision-level loss masking to exclude supervision on low-quality behavior. Experiments with GLM-5.3 on PostTrainBench, a benchmark for autonomous post-training, show that training on RSI-Prime yields a 31.95% relative improvement in overall performance over the original GLM-5.3, highlighting a path toward future RSI systems in which models autonomously sustain their own advancement.
Flow matching models, ODE-based generative models, generate samples by gradually morphing a simple source distribution into a target distribution. However, these models still face challenges in optimally balancing computational efficiency with generation fidelity. While predictor-corrector schemes have been proven to help address this issue in diffusion models effectively, their application to flow matching models remains under-explored. In this work we present a novel training-based predictor-corrector framework for flow matching, which exploits the fact that the ODE can be easily reversed. Our key idea, termed Marginal Matching, is to correct samples at a particular time step in the forward ODE towards a matching marginal distribution of the backward ODE. This scheme proves to reduce a bound on the distance between the data distribution and the learned distribution, motivating two different implementations for the correction function. We show that our proposed method improves sample quality on CIFAR-10 and ImageNet-64, with minimal overhead in computation time.
We study the classical limiting laws of $\varphi(n)/n$ and $σ(n)/n$ over integers and shifted primes $p-1$. Let $μ_G$ denote the shifted-prime totient law and $μ_f$ its image under $x\mapsto-\log x$. The law $μ_G$ has Hausdorff dimension zero, while every Borel set of positive $μ_G$-measure has packing dimension one. All four ratio laws and their logarithmic images have lower and upper Shannon information dimensions satisfying $0.550<\underline{d}_1\le\overline{d}_1<0.571$. The law $μ_G$ gives full mass to absolutely normal Liouville numbers. Both $μ_G$ and $μ_f$ have quantitative Fourier decay. For $μ_f$, we obtain sharp concentration asymptotics and localize maximizing intervals. For $U_T$ uniform on $[0,T]$, $(\log T)\widehat{μ_f}(U_T)$ converges in distribution to a random Euler product with an independent uniform phase. For a uniform square matrix over $\mathbb{F}_p$, the probability of a primitive-root determinant tends, as the matrix size grows, to $c(p)=\frac{\varphi(p-1)}{p-1}\prod_{j\ge1}(1-p^{-j})$. Over primes, $c(p)$ has limiting law $μ_G$. In every fixed reduced progression, the empirical laws of $c(p)$ and $\varphi(p-1)/(p-1)$ over primes $p\le x$ converge to their common progression law with sharp Kolmogorov order $1/\log x$. We also obtain transport estimates in growing progressions. Strongly additive functions with prime values $(\log\ell)^{-A}$ have absolutely continuous integer and shifted-prime laws for every $A>0$, disproving the Erdős--Kátai singularity conjecture for $A>2$. Applications include extremal determinant bounds and certified prime selection without complete factorization of $p-1$. These results are unconditional; further quantitative refinements assume the generalized Riemann hypothesis for Dirichlet $L$-functions.
A Neural Envelope Method for Rapid Gradient-based Optimum Takeoff Trajectory Design
Samuel Sisk, Xiaosong Du
AI 速读 基于摘要
作者提出将物理约束生成对抗网络(physicsGAN)与神经包络耦合,把带约束的非凸起飞轨迹优化转化为无约束凸优化,并在 Airbus A3 Vahana eVTOL 上验证了精度与加速效果。
cs.LG
查看原文摘要
Urban air mobility arises to mitigate ground traffic congestion. Multidisciplinary analysis and optimization (MDAO) has been introduced to reduce the energy consumption, thereby expanding aircraft performance. Surrogate-based MDAO lowers the cost of optimization, but multimodality disrupts global optimum search for both simulation-based and surrogate-based optimizations. Convex envelopes convert multimodal spaces into unimodal spaces, permitting efficient convex optimization; however, convexifying a design space may produce infeasible "optimal" designs due to complex design constraints. To address these challenges, this work proposes the coupling of a physics-constrained generative adversarial networks (physicsGAN) and neural envelope to transform original, constrained, non-convex optimizations into unconstrained, convex optimizations. Specifically, the physicsGAN discovers a feasible design space where all design candidates inherently satisfy all constraints, by penalizing surrogate-predicted constraint violations during training. Moreover, the neural envelope constructs a neural network-based convex envelope for the transformed objective function within the physicsGAN-discovered feasible space. Note that the neural envelope only needs to deal with objective function since constraints are inherently satisfied in the feasible space. The neural envelope is demonstrated on takeoff trajectory optimizations of the Airbus A3 Vahana eVTOL aircraft. Results show that neural envelopes achieve a 99.91% mean accuracy within a flight-condition space compared with reference convex envelopes. Neural-envelope-enabled optimizations achieve 99.84% accuracy on the global optimum design, costing only 0.04 seconds on average. The neural envelope accelerates the trajectory optimization by 53 and 162 times on average compared to the physicsGAN-enabled gradient-based and global optimizations, respectively.
Retrieval-augmented vision-language models (VLMs) answer from two sources of evidence that can each fail: the retrieved text can be wrong or noisy, and the image can be degraded or contradicted. We introduce GIB-200K, 200,000 image-question items built from COCO and Visual Genome, each expanded into controlled stress conditions: hard-negative retrieval noise (1-20 documents), visual ambiguity (Gaussian and motion blur, low-resource crops) and cross-modal conflict with counterfactual flip pairs. Instead of only measuring degradation, we ask how models combine the two modalities. From ideal Bayesian cue combination we derive a testable prediction that yields formal definitions of Knowledge Inertia (text over-weighted relative to the model's own image evidence) and Sensory Capture (the reverse), and we estimate the weights with a measurement-error-corrected regression whose recovery of known weights we verify. A counterfactual flip decomposition separates modality bias from concept priors, and a Grounding Illusion Score measures how much clean accuracy is not grounded in the image. We evaluate 4 open models on one frozen, seeded evaluation split of 2,000 GIB-200K items. All models are accurate on clean inputs (92.5%-94.2%) but adverse accuracy falls to 41.3%-59.9%. The two LLaVA models over-weight the text (Knowledge Inertia, w_T/w_I = 1.229 and 1.322), whereas InternVL3 favours the image (w_T/w_I = 0.680); between 34.8% and 71.4% of the clean answers that are correct are abandoned when the document lies. A training-free consistency check that compares the answer with and without retrieval trades coverage for accuracy under misleading documents, while falling back to the image fails when the image is heavily degraded. All per-instance outputs, code, and an audit that recomputes every reported open-model number are released.
More historical data can improve estimation while introducing relationships that transfer poorly to current market conditions. We therefore examine how training quantity and historical selection interact with forecast horizon in intraday equity prediction. We deploy twelve configurations which combine four sampling and validation policies with 6, 12 and 24 month training quantities under a fixed cross-sectional Transformer and feature set. A monthly walk-forward experiment produces 264 refits and eight cumulative return forecasts from 5 to 120 minutes. Evaluation covers January 2025 through 20 October 2026 and uses cross-sectional Spearman information coefficients against future five-factor residual returns. Six-month recency sampling leads individual models at 5-15 minutes, while multi recency configurations lead at 20-30 minutes and uniform configurations at 60-120 minutes. Increasing training quantity does not uniformly improve accuracy while an equal-weight ensemble of all twelve models exceeds every individual mean at 15, 20 and 25 minutes, but not at the remaining horizons. These findings support evaluating historical selection jointly with forecast horizon without identifying a universal policy or a causal benefit from recency.
GRPO has become the dominant reinforcement learning method for eliciting reasoning capability in LLMs using verifiable rewards. However, it suffers from a fundamental failure mode: when a training question exceeds the base model's current capability, all generated responses fail and receive zero reward, yielding no gradient signal and stalling further capability expansion. An emerging class of GRPO extensions, termed Hint-Augmented GRPO (HA-GRPO), adds a "hint" to the context to reduce the difficulty of the question, thereby producing a usable gradient signal. Nevertheless, existing HA-GRPO methods share a key issue: no hint is available at inference time to append to the question, creating a training-inference context mismatch that leads to inconsistent performance gains. In this work, we reinterpret the function of hinting -- not as reducing question difficulty, but as steering model behavior -- and propose ASH (Activation Steering for Hinting), a novel HA-GRPO method built on this view. During training, the model is presented with both hinted and plain questions, while a trainable steering vector is optimized to widen the margins between the model's generation probabilities for responses to the hinted version relative to the plain version, thereby capturing the question-agnostic component of the behavioral shift induced by the hint. During inference, the trained steering vector is directly injected into the model's activation, reproducing the hint-conditioned behavior without requiring any hint in the context. Experiments with four LLMs (Qwen3-1.7B/4B-Base, Llama-3.2-1B/3B-Instruct) show that ASH consistently improves pass@$k$ on competition-level mathematics benchmarks (AMC, OlympiadBench, AIME24, AIME25) and out-of-distribution benchmarks (Minerva, GPQA-Diamond). Notably, it improves average pass@8 of Qwen3-1.7B-Base and Qwen3-4B-Base by \textbf{2.60}\% and \textbf{2.21}\%.
We investigate how allowing self-loops and multi-edges (i.e. moving from simple graphs to pseudographs) impacts the complexity of graph problems on finitely defined monotone graph classes. We do this by extending the C123-framework beyond simple graphs. We use several testbed problems to illustrate some of the new and varied behaviours introduced by the transition to pseudographs. We show that Pseudograph Matching Cut and Pseudograph d-Cut are C123-problems. Consequently, for every finite set HH of simple graphs, we obtain a complete complexity classification for these problems on HH-subgraph-free graphs. This sharply contrasts with their simple-graph counterparts, which are not C123-problems and their complexity remains open even for classes defined by a single forbidden subgraph H. We also show that Pseudograph Stable Cut is not a C123-problem. Nevertheless, for graph classes defined by a single forbidden subgraph, the problem mirrors the C123 classification; this correspondence already fails for sets of size 2. This gives the first example of a problem that aligns with C123-problems for a single forbidden subgraph but not for larger sets. Motivated by the interpretation of Stable Cut as a surjective homomorphism to a partially reflexive path, we also study surjective homomorphisms to partially reflexive cycles on k vertices with one irreflexive vertex. These problems share structural similarities with Pseudograph Stable Cut. We prove that for k at least 6, they are NP-complete on triangle-free pseudographs. However, unlike Pseudograph Stable Cut, they are polynomial-time solvable on all graph classes with bounded-degree.
Latent Goal Prediction from Language for Model-Based Planning
Samuel Barbeau, Simon Roy, Giovanni Beltrame, Christian Desrosiers, Nicolas Thome
AI 速读 基于摘要
LAGO 用一个共享潜在空间中的单一预测器同时建模动作条件动力学并把语言指令落地为中间潜在子目标序列,从而支持长程基于模型的规划。
cs.AI · cs.LG
查看原文摘要
Joint-Embedding Predictive Architectures (JEPAs) enable agents to plan in latent space by imagining the outcomes of candidate actions, yet task specification remains a bottleneck. Visual targets provide precise local gradients but poor distant guidance, while language is flexible yet limited by noisy cross-modal alignment or dependence on distinct large generative models. We introduce LAGO (Latent Goal Prediction from Language), a hierarchical world model in which a single predictor both forecasts action-conditioned dynamics and grounds language instructions as sequences of intermediate latent subgoals, training both modes with a single regression objective over a shared latent space. At each planning step, LAGO predicts a sequence of latent subgoals from a language instruction and optimizes an action sequence using a soft-minimum alignment cost that rewards subgoal proximity without enforcing a rigid path. Subgoals are repredicted as the agent's state evolves, turning a single long-horizon instruction into a sequence of locally tractable objectives. Across environments spanning navigation and manipulation, LAGO outperforms flat and hierarchical image-goal planners, with the largest gains on long-horizon navigation with curved, hazard-constrained paths, where it more than doubles the success rate of a goal-conditioned behavior cloning policy trained on the same demonstrations. Grounding language in the world model's own latent space also outperforms vision-language reward models, including one that plans with ground-truth dynamics.
Understanding and reconstructing infectious disease transmission remains challenging due to sparsely sampled and biased epidemiological case data: only a small fraction of infectious disease cases are reported to health authorities, and an even smaller fraction are sent to labs for genomic sequencing. In this work, we study the problem of inferring unobserved transmission events by integrating pathogen genomic data with human mobility information. Intuitively, even if we do not have case data for a certain region, such as a census block group, we can use massive-scale individual mobility data to understand where people may be exposed and infer where unobserved cases may be located. Assuming only a subset of cases are reported and sequenced, we reconstruct a phylogenetic tree from observed cases and transform it into a partial transmission tree using the outbreaker2 tool. This tree contains inferred transmissions based on the genetic distance of the observed cases. As these inferred nodes have unknown locations, we propose a mobility-aware belief propagation approach to estimate their locations using human mobility data. Experiments on simulated and real-world genomic surveillance data with biased sampling show that incorporating mobility data improves the reconstruction of inferred transmission trees, reducing cross-entropy relative to the observed distribution by 39.4% and 16.1%, respectively. Results on real-world COVID-19 sequence data show the promise of combining genomic and mobility data for more reliable location inference, with potential implications for outbreak analysis and public health decision-making.
While gravity is often used in mobile robotics as a global constraint for Simultaneous Localization and Mapping (SLAM) algorithms, most estimation methods rely on indirect observation of the gravity vector. In this paper, we propose the use of altimeter measurements to directly observe the direction of the gravity vector. We provide a theoretical evaluation of the estimation noise and validate it through extensive experimental data, along with an evaluation of the impact of intersensor distance on our gravity vector estimates. Experimental results demonstrate that a median error of $0.87^\circ$ on the tilt estimation can be achieved with our system in semi-static scenarios. Furthermore, we demonstrate the potential of this approach by comparing it with state-of-the-art methods based on Inertial Measurement Units (IMUs). We highlight that we achieve similar performance to that of IMU-based systems in highly dynamic scenarios and even outperform in specific cases of accelerometer measurement saturation. Finally, an additional experiment shows the potential use within a robotic system through the integration into an Iterative Closest Point (ICP) algorithm.
Memory-efficient personalization is essential for adapting text-to-image diffusion models while preserving user privacy and operating within the limited computational resources of edge devices. To this end, we propose Mix-Opt, a novel mixed optimization framework that adaptively chooses between backpropagation on low-resolution images (BP-low) and zeroth-order optimization on high-resolution images (ZO-high), guided by the characteristics of the diffusion process. As observed in our experiments, BP-low efficiently adapts the model to target-specific features but suffers from structural distortions due to resolution mismatch. Conversely, ZO-high refines high-resolution details with minimal memory overhead but faces slow convergence when applied without prior adaptation. By complementing both methods, our framework leverages BP-low for effective personalization while using ZO-high to maintain structural consistency, achieving memory-efficient and high-quality fine-tuning. To maximize the efficacy of both BP-low and ZO-high, we introduce a timestep-aware probabilistic function that dynamically selects the appropriate optimization strategy during training. This function mitigates overfitting from BP-low at high timesteps, where structural information is critical, while ensuring that ZO-high is applied more effectively as training progresses. Experimental results across both U-Net- and DiT-based diffusion models demonstrate that Mix-Opt achieves competitive personalization performance while significantly reducing memory consumption. Mix-Opt enables scalable, high-quality personalization without modifying the inference process or increasing inference latency.
Effective multi-agent collaboration requires agents to adapt task allocation as goals and responsibilities evolve. Prior task-allocation frameworks often assume static goals, full observability, or rely on large language models (LLMs) without explicit, verifiable beliefs about task allocation or intent. These assumptions limit their applicability to real-world collaboration, where agents often lack full information about each other's actions, progress, or local environment. To support dynamic task allocation under partial observability, we introduce a communication-augmented Bayesian task-allocation framework. Our agent uses dialogue to reduce uncertainty about the environment and collaborators, while keeping decisions grounded in explicit probabilistic beliefs. We evaluate our framework in an Overcooked domain extended with dynamically arriving tasks under both full and partial object-space observability. Our communication agent consistently outperforms a vanilla LLM agent in both settings. Additionally, our framework performs comparably to a non-communicating task-allocation baseline under full observability, but substantially outperforms it under partial observability as task complexity increases by an average speed up of 0.585 and completes on average 18 timesteps faster. These results indicate that our agent is more efficient in dynamic task allocation than baselines in partially observable environments, and that communication is important as available information decreases.
Auditing a Learned Simulator for Legged Robots: The Robotic World Model on ANYmal D
Joyjeet Singh
AI 速读 基于摘要
作者报告了对 ANYmal D 四足机器人学习型模拟器 RWM 及其不确定性版本 RWM-U 的端到端审计,发现单步检查会误判设计选择,且模拟器精度不能预测策略行走表现。
cs.RO · cs.LG
查看原文摘要
A legged robot can learn to walk inside a learned simulator: a neural network that predicts the robot's next state and is run forward on its own predictions. They are usually trained and checked one step ahead, although a policy uses them for a hundred steps or more. We audit one published system end to end, without a GPU: the dynamics model of the Robotic World Model (RWM) for the ANYmal D quadruped, and the uncertainty estimate of its follow-up, RWM-U. We rebuilt the model to match the released code's outputs, losses and gradients exactly, measured 101 trained simulators, 200 compressed copies and 38 simulators of three other architectures, and trained policies inside simulators that we tested in a hand-built MuJoCo model of the robot. Training on the model's own rollouts beat one-step training at 368 steps on all ten held-out episodes (registered), although one-step training was better one step ahead. One-step checks therefore misjudge design choices: pruning 30% of the weights kept 90% of the model's skill one step ahead and 39% at 2 s. Ensemble disagreement ranked rollout error well (Spearman +0.90 at 100 steps), yet covered 4.6% of outcomes at one sigma, against 68% for a calibrated estimate, as each network's own variance collapses under the implemented loss. Two checks that need no recorded future ranked some pools of simulators well, but neither was shown better than 8-step error in a primary test. At the robot level, accuracy has not predicted walking: all 18 fine-tuned copies we tested predicted better than the released model one and eight steps ahead, yet 13 gave failing policies, and registered tests of whether any check ranks simulators by their policies were not settled. In a secondary test, running the authors' policy inside each simulator ranked them well. We also report nine of our own claims that we withdrew. Scope: one robot, one gait, simulated data, no hardware.
Optimal Min-Entropy Estimation with High Probability
Qisheng Wang
AI 速读 基于摘要
该论文确定了经典分布与量子态在高概率下估计最小熵的最优样本复杂度,并给出简短证明。
cs.IT · math.ST · quant-ph
查看原文摘要
Min-entropy plays a fundamental role in assessing the unpredictability of randomness sources. In particular, NIST SP 800-90B specifies min-entropy estimation as a key procedure for entropy sources used in cryptographic random bit generators, illustrating the practical importance of reliable min-entropy estimation. In this paper, we settle the sample complexity of min-entropy estimation with high probability in both the classical and quantum cases. Specifically, for min-entropy estimation with additive error $\varepsilon$ and probability at least $1-δ$, the sample complexity is $Θ(d\log(d/δ)/\varepsilon^2)$ for classical distributions on $d$ symbols and $Θ((d^2+d\log(1/δ))/\varepsilon^2)$ for quantum states on a $d$-dimensional Hilbert space. Our results admit short and simple proofs.
Identifying whom a person looks at is central to understanding multi-party conversations, yet gaze geometry alone becomes ambiguous as more people appear. Conversations offer cues that geometry lacks, such as who is speaking and what is said, but no gaze dataset provides them for many multi-party scenes. We introduce SORBET, a benchmark of 3,340 samples from 1,578 clips with synchronized speech, transcripts, and attended person labels, covering more multi-party scenes than existing gaze datasets. Prior gaze models struggle to use these cues, which come in different forms, must be linked to the right person, and can conflict with gaze geometry. We propose Nunchi, a framework that constructs a text prompt from the transcript and the outputs of gaze and speaker models, links every cue to the candidates with shared labels, and lets a vision-language model (VLM) weigh gaze geometry against conversational cues to identify the attended person. Nunchi outperforms the strongest existing method by 4.8 percentage points. On SORBET, conversational cues help whereas additional visual cues do not, and the gains hold across three VLMs, with automatic transcripts, and on two existing datasets.
Photographic color grading shapes the tone and color of a photograph to achieve a desired look. Text-driven photographic look transfer lets users specify these adjustments in natural language, with the goal of reproducing the requested look across photographs while preserving their content. However, prior research on text-driven color grading uses different datasets, text instructions, target definitions, and metrics. These differences make it difficult to compare how faithfully the methods reproduce the same look. To compare methods under a shared protocol, we introduce LUTBench. We collect 91 three-dimensional lookup tables (3D LUTs) and define one photographic look per LUT with a specified input conversion. Applying these looks to 5,000 MIT-Adobe FiveK photographs produces 455,000 input-target image pairs with reproducible target images. For each look, we use templates to turn measured tone and color changes on fixed color samples into candidate clauses. We use training image measurements to verify applicable clauses before composing the text description. The evaluation measures target fidelity, tests sensitivity to predefined input conversions, and separates generalization to unseen looks from reproduction of seen looks on new photographs. Across 16 methods, stronger text alignment does not guarantee target fidelity. However, most supervised variants improve fidelity on both seen and unseen looks.
When shown matches between two photographs of the same cathedral, people can tell at a glance which ones are wrong: a window matched to its look-alike jumps against its neighbors, while correct matches shift together. Two-view correspondence learning, also called correspondence pruning, asks a network to make this judgment for thousands of candidate matches, most of which are wrong. Mamba, a fast sequence model, now drives strong pruners, but it reads the matches as a plain sequence, and whether its scan can capture such agreement is unclear. We propose SFMambaNet, which scans the matches cluster by cluster, so that correct matches vary slowly along the sequence while wrong ones stay erratic, and filters the scanned features with learnable frequency-domain gates between its forward and backward passes. On camera pose estimation, outlier removal and homography estimation, we find that SFMambaNet outperforms spatial-domain Mamba baselines with about two million parameters. Contrary to the usual expectation that such gains come from extra capacity, we find that a parameter-free low-pass filter beats point-wise layers as large as the gates. The learned gates turn low-pass on their own and make the scan far less sensitive to how the matches are shuffled. Together, these results show that frequency is a natural language for describing agreement between correspondences, with implications for other sequence models that must read unordered sets, such as point clouds and graphs.
Diffusion and flow models have recently been extended to language modeling, generating text through continuous denoising in token embedding spaces. Pretrained models provide token representations and contextual prediction capabilities, but improving generation with limited additional training requires an appropriate connection between supervision and continuous generation. When continuous predictions are formed by weighting token embeddings with predicted probabilities, different probability distributions can yield the same vector and are therefore indistinguishable under a continuous reconstruction objective based on mean squared error (MSE). We propose Token Identity Brier Flow (TIBF) to apply squared supervision directly to token identities. TIBF fixes token embeddings, combines network scores with token-matching scores under a Gaussian observation model, supervises token probabilities using the Brier loss, and uses probability-weighted embeddings for continuous generation. We show that, for fixed token embeddings, the squared error in token posterior prediction controls the squared errors of the continuous denoising prediction and the flow velocity. A risk decomposition separates uncertainty inherent in the observation from posterior fitting error, motivating continued training that preserves pretrained embeddings and focuses on probability prediction. From common pretrained initializations and under matched continuation budgets, TIBF achieves mean Gen.PPL values of 34.87 on OWT and 80.08 on LM1B, outperforming continuous-MSE and CE controls. Further evaluations show improvements across multiple sampling budgets. Code is available at https://github.com/Jiangjiawei-kust/TIBF.
Event cameras offer high temporal resolution and low power consumption, making them promising for speech-related applications. These include automatic lip reading, audiovisual speech analysis, and human-robot interaction, particularly in challenging environments characterized by low illumination or high dynamic range. Event-based lip reading, speaker recognition, and voice activity detection, however, mostly operate on a cropped region of interest (ROI), by hand or by the dataset, and no event-based method supplies that ROI as a spatial region; existing event-based face analysis gives the mouth as landmarks, which presents a mismatch between application and localization. The present work predicts a dense heatmap over the lip region from the full event "frame". Events are encoded as a polarity-separated voxel grid and processed at every location by a two-pathway continuous-time recurrent network (CT-RNN) with learned, log-space bounded time constants acting on the real elapsed time of each bin: a stateless fast pathway and a stateful slow pathway, utilised for refining and anchoring the prediction. The model reaches 69.5% on lips on four unseen speakers and switches target only 2.1 times per second, against 8.9 and 11.0 for convolutional baselines with more parameters. Adding a spatial encoder improves precision but fails on a new face scale. On a real event sensor, trained only on simulated events, the proposed model reaches 58.0% on lips against 43.2% for the CNN, with fewer than half its target switches.
This paper presents a hybrid neural-symbolic detector of distributed denial-of-service (DDoS) attacks on the availability of operational technology (OT) networks, designed for their resource-constrained gateways, and an evaluation matched to deployment. A compact gated recurrent unit (GRU) is fused at score level with a shallow, readable decision tree, and weight and threshold are selected on validation data under an explicit false-positive budget, whose attainment in finite samples is analysed. The evaluation on CIC-DDoS2019, Edge-IIoTset and CICIoT23 uses contiguous or capture-held-out partitions, dependence-aware intervals, and precision and alert load at operational prevalence. Over ten seeds at a 1% budget, the hybrid's false-negative rate on CIC-DDoS2019 is 0.08%, against 0.26% for the GRU and 0.17% for the tree. With a day of mostly unseen attack families held out, the tree alone carries the decision. Audits bound these claims: Edge-IIoTset is saturated by duplicates, no benchmark as used provides interleaved traffic in capture order, and an unconstrained tree is more accurate on two of them, so the case for the hybrid rests on short, stable rules and a measured footprint. On a Siemens SIMATIC IoT2050 gateway, tested with Modbus/TCP and MQTT floods, the INT8 model needs 2.9 ms per window, 31 MB of memory and 3.5 mJ per inference.
Robust Sunflower lemmas imply that any large enough monotone DNF of width $w$ contains a sunflower, i.e., a DNF equivalent to the conjunction of a common core with a DNF that is heavily biased towards $1$. While these lemmas are typically proved in the context of the uniform distribution or i.i.d. distributions, they also hold for $O(w^2)$-wise independent distributions by Bazzi's theorem (Bazzi (SICOMP 2009); Tal (CCC 2017)), which states that width-$w$ DNFs cannot distinguish between such distributions and the uniform distribution. We prove a derandomization of robust sunflower lemmas for $O(w)$-wise independent distributions, which recovers the derandomization above with slightly weaker parameters. Notably, our result also holds for distributions whose marginals are "locally independent" and "approximately identically" distributed. It is not known whether such distributions fool general DNFs, which makes the derandomization via DNF-fooling techniques unavailable. Our proof of this derandomization is via the method of moments, which might be of independent interest. We apply this result to obtain improved separations between non-monotone arithmetic circuits and the radical of monotone circuits. Specifically, we construct an explicit family of multilinear polynomials $P_n$ that have small non-monotone arithmetic circuits such that any power of $P_n$ cannot be computed by monotone arithmetic circuits of sub-exponential size. This improves on a recent result of Cavalar, Fabris, Mukhopadhyay, Srinivasan and Yehudayoff (STOC 2026), who proved a non-explicit separation that was quasipolynomial. Our construction uses Nisan-Wigderson designs with an additional linear-algebraic expansion property.