Machine Learning
Explore Broadly, Reason Sharply: Push Small Models toward the Frontier via Sampling
Panagiotis Theodoropoulos, Nan Jiang, Xintong Duan et al. (+4 more)
Power-sharpened sampling is an inference-time alternative to reinforcement-learning (RL) post-training for enhancing
reasoning in large
language models (LLMs). High-probability sequences are amplified under the base model without parameter updates or external rewards, avoiding the costly optimization and jagged generalization of RL. However, this approach faces a fundamental exploration–exploitation trade-off, as % strong sharpening restricts exploration, trapping samplers in plausible but incorrect
reasoning trajectories, whereas weak sharpening leaves the answer distribution diffuse. To resolve this trade-off, we introduce Parallel Power Tempering (PPT), instantiating power-sharpened LLM sampling via parallel tempering. Running multiple interacting replicas in parallel at different sharpening levels allows lower-power replicas to explore diverse
reasoning trajectories and higher-power chains to further exploit higher-likelihood responses favored by the sharpened target. Specifically, we tailor \method{} to inference-time sampling by mitigating a truncation bias, identified in prior power samplers, and investigate effective swap strategies under finite
memory and compute budgets. Extensive experimentation shows that \method{} substantially improves single-chain power-sharpened sampling and outperforms RL-post-trained models, producing higher-quality
reasoning traces and even achieving performance comparable to frontier models.
Read the full abstract ▾
All authors Panagiotis Theodoropoulos, Nan Jiang, Xintong Duan, Ali Hasan, Yuriy Nevmyvaka, Evangelos A. Theodorou, Wei Deng
Data Structures and Algorithms
Solving Linear Systems in $\widetilde{O}(mn \log \frac{\kappa}{\epsilon})$ Bit Operations
Jonathan A. Kelner
We give a deterministic
algorithm that solves a nonsingular linear system $Ax=b$, where $A\in\mathbb{R}^{n\times n}$ has $m$ nonzero entries and condition number $\kappa$, to any relative residual tolerance $0<\epsilon\le1/2$ using $\widetilde{O(}mn\log(\kappa/\epsilon))$ bit operations for inputs with logarithmically many bits per entry. For sparse, polynomially conditioned systems with $m=\widetilde{O}(n)$, this gives an $\widetilde{O}(n^2)$
algorithm for any inverse-
polynomial accuracy, improving on the
algorithm of Peng and Vempala, as sharpened by Nie, whose running time in this regime is approximately $O(n^{2.2707})$ with the best current
matrix multiplication exponent, and largely closing a gap between the idealized performance of the conjugate gradient method in exact arithmetic and the running time achievable with finite-precision computation that has persisted for over 70 years. The
algorithm is surprisingly simple. For integer inputs, we apply Dixon's lifting
algorithm to the perturbed system $(A+I/R)x=b$ for a suitable integer $R$. After scaling, the
matrix of this system is $RA+I\equiv I\pmod R$, so its modular inverse is trivial, and each lifting step needs only a sparse
matrix-vector product with $A$ on $O(\log R)$-bit numbers. Fast rational reconstruction then recovers the exact solution of the perturbed system, which is an $\epsilon$-accurate solution of the original one. Normalization and rounding extend the result to fixed-point and floating-point inputs, with floating-point outputs represented using short integer significands and a common encoded exponent. A computable certificate removes the need for prior knowledge of $\kappa$.
Read the full abstract ▾
Information Retrieval
Effective Dense Retrieval using Only In-Context Examples
Nour Jedidi, Abdul Basit Ali, Hang Li et al. (+1 more)
Turning decoder-only large
language models (LLMs) into strong dense retrievers typically requires some form of retriever training. In this paper, we ask whether LLMs can instead be prompted to produce effective representations for dense
retrieval given only a few in-context examples. To answer this, we introduce RICE (Representations from In-Context Examples), a simple "training-free" approach that extracts high-quality dense representations from LLMs. To do so, RICE conditions the LLM on examples that provide a shared context for query and document encoding. Our results demonstrate that RICE
embeddings can substantially improve the accuracy of
prompt-based LLM
embeddings, establishing it as a simple method to build LLM-based dense retrievers that do not require training. We release our code at https://github.com/nourj98/RICE.
Read the full abstract ▾
All authors Nour Jedidi, Abdul Basit Ali, Hang Li, Jimmy Lin
Artificial Intelligence
NeuronEye: Query-Guided Visual Concept Activation for Vision-Language Reasoning
Ruiyu Yan, Bowen Chen, Shaowen Wan et al. (+1 more)
Current
vision-
language models (VLMs) encode visual information in dense hidden states where object identity, spatial layout, and local attributes are implicitly entangled rather than explicitly disentangled, limiting their ability to isolate and modulate the specific visual evidence required by a given
language query. Inspired by sparse population coding and top-down modulation in biological
vision, we introduce NeuronEye, a plug-in framework that constructs a sparse, concept-level
neuron vocabulary from intermediate VLM representations and selectively activates query-relevant visual concepts during inference. NeuronEye decomposes
vision-
token states into an overcomplete sparse basis organized by concept-level clusters, uses the
language query to activate relevant clusters and localize the patches where selected concepts are expressed, and injects the focused evidence back into
vision tokens. A complementary suppression mechanism attenuates dominant perceptual directions to preserve weaker but relevant cues. All operations run in a single forward pass over a frozen VLM backbone. On Qwen2.5-VL-7B, NeuronEye raises CV-Bench overall accuracy by +3.1 with gains of +9.5 on Distance, and improves BLINK Multi-view by +8.3, with similar trends on LLaVA-1.6-7B. These results suggest that sparse
neuron vocabularies can serve not only as post-hoc interpretability tools but also as active interfaces for concept-level visual
reasoning.
Read the full abstract ▾
All authors Ruiyu Yan, Bowen Chen, Shaowen Wan, Lin Zhao
Information Theory · Cross-listed from physics.optics
Receiver-Informed Transmitter for 200Gbps C-Band Direct-Detection PON With FFE-Free ONU
Safiya Dabwan, Geraldo Gomes and Stylianos Sygletos
We introduce and numerically demonstrate receiver-informed transmitter (RITX) waveform design for 200 Gb/s PAM-4 C-band direct-detection PON. Using 50 GHz optoelectronics, RITX achieves N1/E1 loss classes with FFE-Free ONU, enabling 40 km reach and 48.1%/66.7% network-wide complexity savings at 1:64/1:128 splits
Read the full abstract ▾
Machine Learning
Tail-Influence Sampling for CVaR Policy Evaluation
Pauline Bourigault, Xiaotong Ji, Matthieu Zimmer et al. (+2 more)
Policies with similar mean returns can differ sharply in rare failures, yet estimating lower-tail conditional value-at-risk (CVaR) accurately can require many costly rollouts. When different conditional components of a
stochastic workflow can be queried separately, we ask how to allocate a fixed evaluation budget to estimate a fixed policy's CVaR most accurately. We derive a tail influence for each queryable conditional law that aggregates how its uncertainty affects CVaR across every Bellman reuse. Its variance yields the fixed-design efficiency bound and the oracle Neyman allocation. Tail-Influence Sampling (TIS) estimates these influence scales from a pilot model and reallocates fresh queries toward kernels that matter most for the tail; a visitation-anchored variant protects against pilot underallocation. Under fixed dimension and a positive
quantile margin, TIS attains oracle asymptotic variance and first-order MSE including pilot cost, while the anchored variant is within a factor two of the oracle. We also characterize an exact-grid regime in which tail- and mean-optimal allocations coincide. On CliffWalking, TIS reduces MSE by 41% versus learned occupancy and 76% versus complete rollouts at the same charged transition budget. In frozen
language-model review workflows, anchored TIS beats an equally regularized mean-influence blend in 23 of 24 MMLU-Pro settings and reaches 2.4-3.4$\times$ lower MSE than rollouts on six-call FinQA reviews.
Read the full abstract ▾
All authors Pauline Bourigault, Xiaotong Ji, Matthieu Zimmer, Rasul Tutunov, Haitham Bou-Ammar
Machine Learning
Probe-Space Preconditioning for Fast and Stable Zero-Order Training
Francois Chaubard, Mykel J. Kochenderfer and Chris Ré
Backpropagation (BP) dominates deep learning but imposes a massive
memory tax. For example, training OPT-30B with Adam requires $\approx$ 600GB of GPU
memory (assuming batch size 8 and sequence length 2048). Alternatively, zero-order optimization (ZOO) trains in inference-mode (requiring only $\approx$ 60GB for the same model): no stored activations, no gradients, and no optimizer states. However, ZOO convergence has lagged behind BP. In this work, we evaluate two methods to close this gap. First, we show that reallocating training compute budget from many steps to large effective batch sizes with many perturbations (or probes) but fewer steps, allows 1SPSA (Spall, 1992) to outperform zero order methods like MeZO (Malladi et al., 2023) with less training compute. Next, we introduce 1.5-SPSA, adding a single "clean" forward-pass per step to 1SPSA to calculate a cheap diagonal preconditioner in probe-space, which improves convergence rate and convergence by down-weighting high
curvature directions. Benchmarking on 6 post-training
datasets on both Qwen3 and OPT model families, we show that 1.5-SPSA achieves State-of-the-Art results over previous ZOO solvers with much less optimization steps. For example, we train OPT-13B (for direct comparison to MeZO) and find 1.5-SPSA achieves +3.1% accuracy on SST-2 over both MeZO and BP in only 70 steps vs. MeZO's 100,000 steps. Finally, we combine an 8-bit-packing random generator, triton fused unpack/apply kernels, and distributed parallelism to achieve fast and stable training of models as large as OPT-30B in-place on commodity GPUs (e.g. A100).
Read the full abstract ▾
Machine Learning
Dimensionally consistent surrogate modelling through dimensional analysis and harmonic expansions
Ernest Tarrus and Hector Gisbert
Dimensional homogeneity is a fundamental constraint on physically meaningful models, requiring invariance under changes of units. We present a data-driven method for constructing surrogate models that satisfy this constraint at the level of the hypothesis class. Starting from a dimension
matrix of measured variables, the method derives Buckingham $\Pi$-groups, constructs admissible dimensional prefactors, and approximates the remaining dimensionless dependence using truncated harmonic expansions on normalized invariant domains. Once the prefactor and dictionary are fixed, the coefficients are obtained from a regularized linear regression problem. We test the approach on the simple pendulum, Planck's black-body law, the double-pendulum Lyapunov field, and an experimental COBE/FIRAS black-body spectrum
dataset. The results show that dimensional constraints improve conditioning, robustness to noise, and sample efficiency relative to unconstrained baselines, while the choice of dictionary becomes important in non-periodic or multi-invariant settings. The learned expressions are explicit and inexpensive to evaluate, which makes them useful as surrogate models for structured physical problems.
Read the full abstract ▾
Artificial Intelligence
Character Training for Risk-Averse Agents
Arav Dhoot, Punya Syon Pandey, Jamie Johnson et al. (+3 more)
Risk aversion in resources could prevent misaligned AI
agents from causing catastrophic harm. Misaligned but risk-averse
agents would tend to favor safer strategies like making deals with humans over riskier strategies like rebelling. We train
agents to be risk averse through character training, finding that persona traits provide a robust mechanism for instilling risk preferences. To do this, we construct a model constitution describing constant absolute risk aversion (CARA) over an
agent's resources and instill it through on-policy
distillation. Despite never seeing the
benchmark's decision format during training, character-trained models are competitive with baselines trained directly on it, and generalise better than them out of distribution on two of our four models. We also modulate different aspects of the constitution, finding that
token budget and model choice are the most influential aspect of character training to instill risk aversion. We conclude from these results that character training is a promising and scalable way to instil broad dispositions, which we can use to our advantage in mitigating risk from misaligned AI
agents.
Read the full abstract ▾
All authors Arav Dhoot, Punya Syon Pandey, Jamie Johnson, Daniel Tan, Elliott Thornley, David Demitri Africa
Machine Learning
Mira: Memory-Efficient MoE Inference Using Adaptive Caching and Predictive Expert Staging
Sanjali Yadav and Bahar Asgari
Mixture-of-Experts (MoE) models are a compelling architecture for scaling model capacity, making them especially attractive for deployment on resource-constrained, single-GPU systems. However, this benefit is difficult to realize because expert parameters dominate
memory, and
token-level routing is dynamic, unpredictable, and skewed. Prior work using offloading and caching remains fundamentally reactive, as systems wait for router outputs before moving experts, leading to inefficient cache utilization and an inability to overlap transfers with compute under tight VRAM budgets. To address these challenges, we propose Mira, an
algorithm-system co-design that enables high-capacity MoE inference on a single GPU. Mira shifts from a reactive to a proactive stance by coupling predictive expert management with a tailored quantization format. It introduces lightweight per-layer predictors that anticipate expert usage two layers ahead, enabling proactive prefetching. These predictions feed a two-tier HOT+STAGE GPU cache managed by
token-level routing telemetry to retain frequently used experts while staging predicted ones. To minimize transfer overhead, Mira implements a custom compression for expert parameters, which reduces metadata and improves packing efficiency, while minimally degrading accuracy. Mira is implemented as a fully integrated runtime that coordinates predictors, caching policies, and quantized transfers to maximize overlap between communication and compute. Our experiments show that Mira reduces expert-induced stalls. Compared against state-of-the-art baselines, Mira achieves a 5.71x speedup in average throughput on a
memory-constrained GPU. It accelerates Time-to-First-
Token by 11.71x and achieves a 3.84$x average speedup in beam search inference, demonstrating its effectiveness across diverse inference scenarios.
Read the full abstract ▾
Computational Engineering
Neural topology optimization of ship structures under propulsion machinery vibrations
Shengyu Yan, Muhammad Muztahidul Hakim Zareer and Jasmin Jelovica
Ship structural vibrations contribute to noise, fatigue, and equipment damage, while dynamic-compliance
topology optimization can produce pathological designs near resonance. This study extends neural-reparameterized
topology optimization using a convolutional Kolmogorov-Arnold network (KATO) to forced-vibration design with active input power (AIP) as the objective. Applications include a 100 Hz engine-supporting deck panel and an 18 Hz thruster foundation frame. Helmholtz PDE filtering and Heaviside projection control feature sizes and manufacturing tolerance. Across both deck families, all eight optimized layouts reduce AIP relative to size-optimized references and, after finite-depth extrusion, also achieve lower static compliance. For unrestricted, manufacturing-aware, and stress-aware frame variants, KATO matches GCMMA in AIP within 0.5 dB while yielding 22-36x lower static compliance after matched-volume binary re-analysis. In a near-resonant 300 Hz case, both methods reduce initial AIP by more than 32 dB; KATO maintains a connected design, achieves 59x lower binary static compliance, and reduces maximum AIP over 1-500 Hz by 2.7 dB. KATO runs 6.4-10.4x faster than GCMMA for the implemented stress-aware formulations. The results demonstrate neural AIP-driven
topology optimization as an efficient approach for designing connected, feature-size-controlled ship structures with improved forced-vibration performance.
Read the full abstract ▾
Robotics
CrossBFM: Distilling a Shared Latent Behavior Space Across Humanoid Embodiments
Tan-Dzung Do, Tuan Dat Phuong, Nico Bohlinger et al. (+6 more)
Behavior Foundation Models (BFMs) give
humanoids a promptable policy over a latent behavior space, enabling one single vector to represent a motion to imitate, a pose to reach, or a reward to maximize. Forward-Backward representations successfully produce such spaces, but at the cost of hundreds of GPU-hours for a single
robot. Moreover, when the training process is repeated for a second
robot, it produces a second space unrelated to the first, resulting in embodiment-specific latents that do not unify or transfer. We address these problems with CrossBFM, treating the latent space as the transferable asset for various embodiments. As retargeting provides frame-level cross-embodiment correspondence, we propose a unified encoder architecture with no
robot-specific parameters for distilling the behavior space to address all training embodiments simultaneously in less than a GPU-hour. Following this encoder, latent-conditioned trackers turn the distilled latent into whole-body control in a conventional PPO training manner in just 10 more GPU-hours. On three distilled
humanoids, all three prompting modes transfer: motion tracking with latent-conditioned policy losing only $0.025$ rad to its joint-conditioned counterpart, smooth goal reaching between poses with no falls, and reward optimization for all $41$ reward
prompts. Our experiments further reveal that 1) regressing the encoder on a quarter of the motion corpus costs only $5\%$ of tracking performance and 2) training the encoder on a subset of
robots and evaluating on an unseen one recovers up to $89\%$ of the tracking performance of seen
robots, demonstrating cross-embodiment generalization to morphologically similar
robots. We also verify the pipeline on real
robots across all three prompting modes and with flow-based generated latents. Project website: https://dotandung.github.io/crossbfm/
Read the full abstract ▾
All authors Tan-Dzung Do, Tuan Dat Phuong, Nico Bohlinger, Cuc T. Trinh, Siwei Ju, Vien Anh Ngo, Jan Peters, Xinchao Wang, An T. Le
Computer Vision
VISTA: Internalizing Collective Visual Experience via On-Policy Distillation for Active Multimodal Agents
Zheng Jiang, Houde Qian, Yiming Chen et al. (+5 more)
Active
multimodal agents use visual tools to acquire task-relevant evidence while
reasoning. Although reinforcement learning samples multiple interaction trajectories per input, outcome-based objectives primarily use the group to estimate scalar advantages, leaving complementary visual discoveries underused. We introduce VISTA, which internalizes collective visual experience through on-policy
distillation by turning observations from same-input rollouts into shared supervision. Collective visual experience
distillation (CVED) organizes these observations with their interaction context and aligns them with individual decisions, while heterogeneity-aware policy improvement (HAPI) reinforces successful trajectories and provides experience-guided
distillation for unsuccessful attempts. An experience-conditioned teacher evaluates the student's sampled response prefixes, allowing discoveries from one trajectory to guide learning in another without replacing the student's original history or generating new target trajectories. The trained
agent retains its visual tools and acts using its own interaction history. VISTA achieves the strongest average performance among the evaluated active
multimodal agents of comparable size and consistently outperforms same-backbone training baselines across fine-grained perception and general
reasoning tasks, demonstrating the value of collective experience for active
multimodal learning.
Read the full abstract ▾
All authors Zheng Jiang, Houde Qian, Yiming Chen, Ling Li, Chaoyang Li, Yueqi Li, Yuxuan Liu, Lifeng Sun
Machine Learning
Traversing the solution space of neural networks with Hessian Null Space Continuation
Ann Huang, Mitchell Ostrow, Zhouyang Lu et al. (+3 more)
On a single task, deep networks can learn many solutions, depending on their optimizer, training data, architecture, and hyperparameters. Many of these solutions are mode-connected: rather than isolated points in weight space, they are connected by low-loss regions. Yet how their internal computation varies within these regions is unknown. A parallel line of work has identified the degeneracy of neural representations: many networks reach similar training loss with distinct internal structures. However, it is unclear how these solutions are related in weight space. We unify these subfields and show for the first time that many different internal mechanisms exist within a local mode-connected region in weight space. To do so, we introduce Hessian Null Space Continuation (HNC), a scalable method that uses local
curvature to traverse regions of weight space that preserve network function, and can be steered toward solutions with specified properties. In RNNs trained on a
memory task, HNC reaches drastically different representations and dynamics with maintained behavior. In ImageNet-trained
Vision Transformers, HNC finds representations that differ more from the original network than any independently trained model with a different architecture or objective. In reinforcement-learning
agents, HNC uncovers a distinct
navigation strategy at comparable return and exposes reward hacking in an AI Safety Gridworld. Finally, HNC measures the local geometry of the solution set, showing how model size and task complexity shape its dimension and functional sensitivity. Our results show that a surprisingly large amount of representational diversity exists near a single trained solution, unseen by standard gradient-based optimization. HNC identifies and quantifies this diversity, opening new possibilities for mechanistic understanding of solution spaces and for model merging, editing, and fine-tuning.
Read the full abstract ▾
All authors Ann Huang, Mitchell Ostrow, Zhouyang Lu, William T. Redman, Leo Kozachkov, Kanaka Rajan
Computer Vision
OmniTaskonomy: When Does Visual Generation Improve Visual Understanding?
Jiaxin Ge, Yiming Qin, Ji Xie et al. (+13 more)
Training a model to generate visual content can encourage it to learn rich perceptual capabilities related to geometry, spatial relationships, and objectness; yet, its benefits for visual understanding remain unclear. We ask: when and how does visual generation supervision improve visual understanding? We study controlled pairs of image-to-image (I2I) generation and image-to-text (I2T) understanding tasks that express the same underlying problem in different output modalities. We find that under the correct recipe, I2I training improves downstream I2T performance, with larger gains as the amount of I2I training data increases. We next ask which generation tasks benefit which understanding capabilities. To study transfer beyond paired tasks, we introduce OmniTaskonomy, a unified taxonomy spanning 19 I2I generation tasks and 25 I2T understanding capabilities. The resulting transfer map reveals selective, task-dependent benefits. Some follow intuitive correspondences, e.g., depth prediction improving metric 3D
reasoning, object pointing improving counting, and jigsaw reconstruction improving 2D ordering. Interestingly, we also uncover surprising connections: 2.5D
segmentation improving category recognition and Z-depth prediction improving localization. To probe these patterns, we analyze gradient
alignment between generation and understanding tasks and find that stronger
alignment is associated with larger downstream transfer gains. Together, our results highlight visual generation as a rich source of supervision for visual understanding and provide a roadmap for unlocking its benefits through the right training curriculum and task selection. Project page: https://omni-taskonomy.github.io/.
Read the full abstract ▾
All authors Jiaxin Ge, Yiming Qin, Ji Xie, Haozhe Jiang, Xiaochuang Han, Junyi Zhang, Andrew Dai, Yinfei Yang, Jitendra Malik, Ranjay Krishna, Sewon Min, Haiwen Feng, Le Xue, Baifeng Shi, Trevor Darrell, XuDong Wang
Robotics
MotorMind: Scaffolding General Vision Language Models for Zero-Shot Robot Manipulation
Bingxuan Li, Siqi Song, Yizhuo Wu et al. (+3 more)
Vision-
language-action (VLA) models have advanced robotic manipulation, but their zero-shot generalization in new tasks and environments remains limited, and their reliance on specialized training keeps them from benefiting directly from rapidly advancing general-purpose
vision-
language models (VLMs). In parallel, recent agentic robotic systems leverage VLMs for high-level
reasoning or coding
agents for
robot control, but often depend on extensive external models and tools, introducing additional complexity and cost. This motivates us to ask: Can a general-purpose VLM itself operate a
robot more like the human teleoperator by
reasoning directly from observations, issuing actions, and continuously adapting to execution feedback, without relying on external models such as learned action experts, coding
agents or grounding tools like SAM3? In this work, we introduce MotorMind, a
robot manipulation harness that connects VLM-proposed mid-level actions to deterministic
robot control and feedback, with asynchronous monitoring and background
memory updates. Without task-specific policy training, coding
agents, or additional grounding tools such as SAM3, MotorMind achieves 66.7% success on the base LIBERO-PRO suites and 53.8% under perturbations, compared with at most 13.3% and 19.2%, respectively, for the prior zero-shot methods we evaluate. The same interface reaches 95% average success on a real xArm6
robot across direct manipulation and human-perturbation settings. Replacing the backbone with a stronger VLM further improves performance, while the remaining failures - primarily due to visual grounding,
embodied reasoning, and action knowledge - decrease as VLM capability improves. These results show that a general-purpose VLM, when equipped with an appropriate mid-level action representation and asynchronous execution harness, can perform effective zero-shot robotic manipulation.
Read the full abstract ▾
All authors Bingxuan Li, Siqi Song, Yizhuo Wu, Jiarui Yao, Tong Zhang, Huan Zhang
Computer Vision
MUGEN: Interactive Panoramic World Exploration via Camera Control
Jiaming Tan, Zhen Li, Shuwei Shi et al. (+6 more)
Interactive panoramic
video generation aims to synthesize immersive 360\textdegree{}
videos that remain visually coherent while following user-specified camera trajectories during exploration. However, progress is limited by a coupled data-and-model gap: existing panoramic
video datasets are often short, weakly annotated, or lack camera trajectories, while existing camera-controlled
video generation models are designed for perspective
videos and do not directly support panoramic geometry. In this paper, we introduce MUGEN and Wan360 to address these limitations. MUGEN is a large-scale real-world panoramic
video dataset tailored to interactive 360-degree world exploration, comprising over 1,300 hours of at least 4K panoramic
videos with rich semantic and
geometric annotations. Built on MUGEN, we further present Wan360, a camera-controllable interactive panoramic
video generation model. Panoramic
videos are commonly represented by EquiRectangular Projection (ERP), which unfolds a spherical 360-degree view into a rectangular frame with cyclic longitude seams and pole distortions. To this end, Wan360 introduces three parameter-free ERP-aware components: periodic longitude RoPE for seam-consistent positional encoding, ERP-aware padding for reducing boundary artifacts, and random roll yaw for consistent learning. For camera control, Wan360 uses a panoramic Plücker
embedding that represents camera motion with ERP rays rather than perspective pinhole rays. Experiments show that MUGEN serves as a data foundation for panoramic world exploration, and that Wan360 enables high-quality, temporally coherent, camera-controllable 360-degree
video generation.
Read the full abstract ▾
All authors Jiaming Tan, Zhen Li, Shuwei Shi, Minggui Teng, Siqi Yang, Yuwei Wu, Bo Zheng, Chuanhao Li, Kaipeng Zhang
Computational Complexity · Cross-listed from quant-ph
Ramanujan quantum expanders from the Weil representation
Siddhartha Jain
For every odd
prime power $q$ and $D=q+1$, we construct an infinite family of Ramanujan
quantum expanders of degree $D$. The construction transfers Morgenstern's Ramanujan Cayley graphs on $\operatorname{PSL}_2(\mathbb{F}_{p})$ for $p$ which is an even power of $q$, through the odd irreducible subrepresentation of the Weil representation of $\operatorname{SL}_2(\mathbb{F}_{p})$. For a
quantum expander of dimension $N$, our implementation uses $O(\log^2 N)$ elementary gates, and $O(\log N)$ ancilla qudits, using a fixed finite gate set depending on $q$. An advantage compared to the previous work of Iyer, Jain, Jordan, and Somma (FOCS 2026) is that assuming the
quantum circuit is implemented exactly, we satisfy the $2\sqrt{D-1}/D$ singular value bound exactly without any additive error.
Read the full abstract ▾
Computational Complexity · Cross-listed from quant-ph
Optimal Quantum-Classical Separations for Exact Learning
Srinivasan Arunachalam, Amin Shiraz Gilani and Nikhil S. Mande
We study exact learning with membership queries for concept classes $\mathcal C\subseteq\{0,1\}^N$, focusing on the relationships among their deterministic, randomized, and
quantum query complexities, denoted $\mathsf{D}(\mathcal C)$, $\mathsf{R}(\mathcal C)$, and $\mathsf{Q}(\mathcal C)$, respectively. The two canonical
quantum speedups in this model are witnessed by Grover search and Bernstein-Vazirani, leading to the longstanding
conjecture $$ \mathsf{R}(\mathcal C)=O(\mathsf{Q}(\mathcal C)^2+\mathsf{Q}(\mathcal C)\log N). $$ We first refute this
conjecture by constructing concept classes $\mathcal C$ and $\mathcal C'$ satisfying \[ \mathsf{R}(\mathcal C)=\Omega\!\left(\frac{\mathsf{Q}(\mathcal C)^3\log N}{\log \mathsf{Q}(\mathcal C)}\right) \qquad\text{and}\qquad \mathsf{D}(\mathcal C')=\Omega(\mathsf{Q}(\mathcal C')^3\log N). \] The first bound matches the upper bound of Arunachalam et al.~[
Quantum'21] up to constant factors, while the second matches the upper bound of Servedio and Gortler~[SICOMP'04]. In particular, this shows that the saving in the randomized upper bound of Arunachalam et al. fundamentally relies on randomness. Apart from characterizing the optimal relationship between classical and
quantum query complexity, our results are the first to show that
quantum speedups for learning can go beyond the Grover and Bernstein-Vazirani paradigms.
Read the full abstract ▾
Computer Vision
RS-OPSD: Reliable Privileged On-Policy-Self-Distillation for Ultra-High-Resolution Remote Sensing VQA
Chengjie Jiang, Yunqi Zhou, Jiafeng Yan et al. (+3 more)
Ultra-high-resolution (UHR) remote sensing visual question answering (VQA) requires models to resolve small visual evidence within extremely large images. Existing approaches typically rely on
token pruning, visual search, or tool-augmented
reasoning at inference time. We instead investigate whether the benefit of zoom-in visual privilege can be internalized into the model. We introduce RS-OPSD, a reliable privileged on-policy self-
distillation (OPSD) framework for UHR remote sensing VQA. To provide high-quality privileged information with explicit question-relevant evidence, we construct GeoEvidence-6K, containing 6,750 VQA samples across seven task categories with evidence-region annotations, and develop Human Feedback-Guided Skill Refinement (HF-SR) for scalable annotation. To address context loss from tight crops and conflicting signals from imperfect teachers, RS-OPSD introduces Context-Preserving Visual Privilege (CPVP) and Correctness-Aligned
Distillation (CAD). Without any additional visual search or tool calls at inference time, RS-OPSD achieves state-of-the-art (SOTA) performance on XLRS-Bench, MME-RealWorld-RS, and LRS-VQA, outperforming previous SOTA models of comparable scale by an average of 4.0 percentage points. Moreover, our 2B variant, RS-OPD-Lite, surpasses most 8B-scale models while achieving the fastest measured inference speed. Our Code, GeoEvidence-6K, and the model weights for RS-OPSD and RS-OPD-Lite are publicly available.
Read the full abstract ▾
All authors Chengjie Jiang, Yunqi Zhou, Jiafeng Yan, Sihang Zhao, Chun Yuan, Jing Li
Artificial Intelligence
Probability is Not Enough: Exploring and Counting Divergent Tokens for Reasoning Uncertainty Quantification in LLMs
Feiyang Li, Shengjing Liu, Qi Zhan et al. (+7 more)
As the chain-of-thought
reasoning capabilities of large
language models improve, evaluating and calibrating their
reasoning confidence is becoming increasingly important for quantifying the uncertainty of their answers. Current methods for estimating the confidence of large
language models are generally based on probabilities of selected key
tokens, but the underlying mechanism remains unclear. Our pilot study finds that replacing selected
token probabilities with coarse substitutes can also improve calibration, motivating us to further explore effective signals of model confidence. We introduce Divergent
Token Confidence (DTC), a framework that estimates confidence by counting
tokens at which two models strongly disagree during decoding. DTC identifies these divergent
tokens using the Jensen-Shannon divergence between next-
token distributions evaluated along the same
reasoning trajectory. We find that their count is almost negatively associated with answer accuracy, thereby serving as a simple yet effective signal for uncertainty quantification. DTC supports both white-box and black-box evaluation using auxiliary models, without explicit training and affecting the generation process. Experiments across multiple model families and six mathematical
benchmarks demonstrate improved calibration over probability-based and verbalized baselines. Under white-box evaluation, the count-only
estimator achieves an average expected calibration error of 13.0%, compared with 32.7%-42.4% for standard full-sequence confidence methods. In black-box settings, it also improves calibration over the original verbalized scores. For example, mean expected calibration error falls from 32.1%-40.2% to 13.7%-16.3% on DeepSeek-V3.2. These findings provide new insights for improving
reasoning uncertainty quantification in large
language models. The code is released at https://github.com/szu-tera/DTC.git.
Read the full abstract ▾
All authors Feiyang Li, Shengjing Liu, Qi Zhan, Sijie Cheng, Weiqing Wang, Hongwen Chen, Yuxuan Yang, Wen Wang, Yile Wang, Hui Huang
Systems and Control · Cross-listed from math.OC
Stochastic Lie-Bracket Approximations for Zeroth-Order Optimization on Manifolds
Mahmoud Abdelgalil, Miroslav Krstic and Jorge I. Poveda
We propose a continuous-time
algorithm for zeroth-order global optimization of smooth functions on compact
manifolds that synthesizes Riemannian gradient dynamics using only objective measurements, without evaluating gradients. The
algorithm is formulated as an ordinary differential equation driven by piecewise-smooth McShane-type approximations of the Wiener process, with a
geometric compensation term that depends only on the actuation frame and not on the objective function. Our main result establishes infinite-horizon practical probabilistic guarantees for fixed approximation parameters. To this end, we show that the
algorithm's trajectories approximate, in the mean-square sense, a Riemannian Langevin
diffusion whose unique invariant Gibbs measure concentrates on the set of global minimizers. We exploit this approximation to establish finite-horizon probabilistic convergence guarantees and directly analyze the
algorithm's long-time behavior to obtain the infinite-horizon result.
Read the full abstract ▾
Machine Learning
A foundation model for energy and radiation systems built on heterogeneous scientific interfaces
Samrendra Roy, Tapas Tripura, Yoon Pyo Lee et al. (+2 more)
Scientific foundation models are commonly evaluated after heterogeneous physical problems have already been translated into a compatible gridded, tokenized or symbolic representation. This leaves the scientific interface outside both the pretrained model and the audit of what is actually reused. We study the complementary setting in which boundary histories, sparse monitor records and loading histories retain their native inference classes and their outputs remain on Cartesian, latitude-longitude and unstructured domains. GEODE couples task-specific scientific interfaces to a shared routed library of wavelet operators. A single jointly pretrained model represents cavity flow, radiation dose and elastoplastic stress, then acquires a heat exchanger and a reactor subchannel by training a private interface containing 2.1% of its parameters. Earlier predictions remain unchanged by parameter isolation, whereas unrestricted fine-tuning degrades them by factors of 14-29. Crucially, preservation alone does not establish reuse: norm-matched randomized-library controls show that the contribution of pretrained computation is conditional on the task and data regime. A separate decomposition shows that full-field relative L2 error can substantially understate error relative to spatial variation when field level dominates the norm. Task-specific operators remain more accurate on three of the five problems. These results distinguish multi-task coverage, preservation and pretrained reuse as separate properties that must be tested independently when scientific foundation models span heterogeneous interfaces.
Read the full abstract ▾
All authors Samrendra Roy, Tapas Tripura, Yoon Pyo Lee, Souvik Chakraborty, Syed Bahauddin Alam
Machine Learning
Alpha Diffusion Language Models: Factorization Alone Is Not the Problem
Nikita Gushchin, Dmitry Baranchuk and Alexander Korotin
Discrete
diffusion language models can generate multiple
tokens in parallel, but reducing the number of denoising steps can lead to inconsistent predictions. Standard cross-
entropy training fits conditional
token marginals, whereas parallel generation requires consistent joint predictions. We introduce Alpha
Diffusion Language Models (AlphaDLM), trained with a sequence-level alpha loss that recovers cross-
entropy in the limit of vanishing alpha and has a joint-mode optimum at alpha one. Our analysis characterizes how the objective and factorization jointly determine the fitted distribution. We identify conditions under which intermediate alpha preserves multiple valid completions while excluding invalid
token combinations. Trained on TinyGSM, our method achieves 34.6% accuracy on GSM8K with only four model evaluations. We further scale the method to SDAR-1.7B and evaluate it on code and mathematics
benchmarks. These results show that changing the training objective can improve the accuracy-computation trade-off of factorized
diffusion language models.
Read the full abstract ▾
Machine Learning
Jaxolotl: A Unified High-Performance Benchmark Suite for LTL-Based Multi-Task RL
Mathias Jackermeier, Jacques Cloete and Alessandro Abate
Training
agents to follow arbitrary instructions is an important goal of multi-task reinforcement learning (RL). Linear temporal logic (LTL) provides a precise and structured formalism for specifying instructions to
agents, and has been successfully adopted for training generalist multi-task policies. However, differences in implementations, task distributions, and evaluation protocols make existing methods difficult to compare, while high computational costs limit the scale and statistical reliability of experiments. We introduce Jaxolotl, a unified high-performance
benchmark suite for multi-task LTL-RL to address these concerns. Jaxolotl provides a modular, end-to-end JAX implementation of six representative
algorithms and four environments, together with newly curated task suites and a standardised, statistically robust evaluation protocol. By precompiling symbolic task representations into static arrays, Jaxolotl enables fully JIT-compiled training and evaluation, achieving end-to-end speedups of up to $220\times$ and supporting controlled comparisons at substantially greater experimental scale. We use this framework to systematically evaluate existing approaches, revealing complementary strengths and limitations: general methods capable of non-myopic
reasoning struggle as the number of propositions grows, while methods with stronger scaling rely on environment-specific assumptions and suffer from myopia.
Read the full abstract ▾
Cryptography and Security · Cross-listed from quant-ph
Classical Verification of Quantum Computation with Quasilinear Resources, from Compiled Nonlocal Games
Finn Holler and Anand Natarajan
Computational self-testing gives a classical verifier command over the
quantum register of a single computationally bounded prover. We use this framework to construct the first argument system for BQP with quasilinear total resource requirements in the
circuit model. Our argument system is based on the learning with errors (LWE) assumption and requires total resources of $O(\mathrm{poly}(\lambda, \log g)\cdot g)$ for delegating a
circuit with $g$ gates, where $\lambda$ is the LWE security parameter. This is achieved by constructing a new computational self-test for certifying the prover's
quantum state and using it to dequantize the efficient verification protocol of Broadbent (ToC 2018). Specifically, this self-test enables the verifiable, random remote state preparation of tensor product states of the single-
qubit Clifford observables $\sigma_X, \sigma_Y, \sigma_Z, (\sigma_Y-\sigma_X)/\sqrt{2}$ and $(\sigma_Y+\sigma_X)/\sqrt{2}$, with constant robustness: the verification error is independent of the number of prepared
qubits. This approach was first proposed by Coladangelo et al. (ToC 2024) in the multi-prover setting. We replicate their result in the single-prover setting by applying the compiler proposed by Kalai et al. (STOC 2023)—which turns any nonlocal game into a single-prover argument system—to a modified version of their self-test.
Read the full abstract ▾
Robotics
WorldLine: Action-Driven Visual Simulation for Robotic Manipulation
Shenghe Zheng, Wenbo Li, Jiyao Zhang et al. (+4 more)
Real-world
robot learning is constrained by the cost of collecting experience and evaluating candidate behaviors.
Video generation models offer a scalable foundation for visual simulators that predict action outcomes before physical execution. Yet they often favor visual plausibility over accurate action following and coherent
robot–object dynamics, while action-conditioned simulators depend on scarce, embodiment-specific data that are difficult to share across incompatible control spaces. We introduce WorldLine, an action-driven visual simulator that decouples transferable dynamics learning from heterogeneous action grounding. WorldLine learns manipulation dynamics from more than 10,000 hours of action-free
robot videos and grounds them using over 2,000 hours of action trajectories across more than ten embodiments. An image-space action representation provides a shared control interface across embodiments, while multi-view and failure-enriched training with relational regularization improves interaction-sensitive prediction.
Robot-focused few-step
distillation enables efficient causal rollout while preserving action-critical motion. Across held-out and out-of-domain settings, WorldLine maintains strong visual quality and
robot-motion agreement; on failed trajectories, it improves
robot-mask IoU by 0.1626 over the strongest baseline. It predicts trajectory success with 74% mean accuracy across RoboTwin and AgiBot, one percentage point above the strongest baseline. Without RoboTwin training or adaptation, its rollouts improve task success by up to 21.4 percentage points over direct policy execution. Together, these capabilities make WorldLine a scalable and efficient visual simulator for policy evaluation and
embodied planning. More results are available at \href{https://zhengsh123.github.io/WorldLine/}{project page}.
Read the full abstract ▾
All authors Shenghe Zheng, Wenbo Li, Jiyao Zhang, Bin Xia, Haoyang Huang, Nan Duan, Jiaya Jia
Machine Learning · Cross-listed from stat.ML
Latent Inference-Time Guidance of Time Series Foundation Models
Chloé Hashimoto-Cullen, Amaury Durand, Laurent Bozzi et al. (+3 more)
Time Series Foundation Models (TSFMs) currently provide state-of-the-art results in
forecasting tasks. They are available out-of-the-box and rely on in-context learning to make their predictions, which makes the quality of their performance highly sensitive to the user-selected lookback,
covariates, horizon and training data distributions. In practise, the quality of the forecasts are variable but complementary, which highlights the need for a principled ensembling approach, rather than selecting the best context. This paper introduces Latent Inference-Time Guidance for TSFMs, which adaptively combines a pool of TSFM forecasts through a time-dependent latent space with independent components. The framework comes equipped with identifiability and reconstruction guarantees, whilst maintaining the off-the-shelf aspect of foundation models. We provide experiments on
datasets at various frequencies and from multiple domains: these show that the approach is competitive with traditional ensembling approaches.
Read the full abstract ▾
All authors Chloé Hashimoto-Cullen, Amaury Durand, Laurent Bozzi, Benjamin Guedj, Yannig Goude, Sylvain Le Corff
Computer Vision
EVO-WAM: Evolving World Action Models through Video-Action Verification
Shiyang Zhou, Xionghao Wu, Wenbo Li et al. (+11 more)
Improving
robot policies on new tasks without collecting additional expert demonstrations remains a central challenge in
robot learning. World action models (WAMs) use broad
video priors to jointly predict future
videos and actions, offering a potential source of supervision for adapting to new tasks. However, generated
videos may fail to depict task completion, and even visually successful
videos may be paired with inconsistent actions that lead to execution failure. We propose EVO-WAM, a framework that adapts WAMs to unseen tasks by learning from their own generated
video-action trajectories, without executing candidate actions in an external environment. First, we augment WAM training with state prediction and anchored multi-frame context to enable complete autoregressive rollouts without external execution feedback. Second, we identify reliable training experience by selecting task-completing prefixes with a
vision-
language model and verifying their
video-action consistency with an inverse dynamics model. Third, we iteratively train the WAM on verified prefixes and generate new rollouts with the updated model. On seven unseen RoboTwin 2.0 tasks, EVO-WAM increases average success rates from 26.9% to 68.0% for Cosmos3 and from 28.5% to 46.4% for DreamZero, reaching approximately $2.5\times$ and $1.6\times$ their initial success rates. On three unseen long-horizon composite tasks in the real world, it improves Cosmos3's average success rate from 20.0% to 76.7%, a gain of 56.7 percentage points. Project Page: https://evo-wam.github.io/.
Read the full abstract ▾
All authors Shiyang Zhou, Xionghao Wu, Wenbo Li, Shenghe Zheng, Jiyao Zhang, Songsong Yu, Yijun Yang, Jianhui Liu, Haoze Sun, Senqiao Yang, Li Jiang, Jingyong Su, Haoyang Huang, Zhuotao Tian
Computational Engineering
Active Liquidity On Chain: Evidence from PropAMMs Across Chains
Ozan Solmaz, Lioba Heimbach and Jason Milionis
The liquidity providers of most typical automated
market makers (AMMs) are passive and known to suffer from adverse selection. AMMs traditionally rely on trades executed on them to sync the price with external
markets (and a notion of fair value thereof). As a result, when an external fair value moves, arbitrageurs trade on AMMs to pick off their stale quotes. A recent approach termed proprietary automated
market makers (propAMMs) emerged as a response in 2024: these on-chain programs instead have their singular operator quote from its own inventory, repricing without a trade via efficiently provided price updates. By 2026, propAMMs accounted for more than half of SOL/USDC volume on Solana. We present the first year-long longitudinal measurement of propAMMs on Solana, Base and Monad, and decode the on-chain logic of the dominant propAMM on Base. We find that two seconds after a fill, propAMMs earn 0.37 bps on Solana and 1.19 bps on Base, while AMMs lose 0.22 and 0.62 bps. We empirically quantify and classify their edge as stemming from four factors: propAMMs continuously reprice rather than waiting for a trade, they charge for the source-dependent risk each counterparty might bring, they avoid cross-venue arbitrage from other on-chain
markets (such as other AMMs), and they spoof by filling trades at a worse price than they quote. Finally, we show that on our proxy for retail flow (i.e., fills that arrive when the reference price is not moving) propAMMs collect 0.26 bps on Solana where AMMs collect 2.59 bps, indicating that they offer better prices for short-term uninformed flow.
Read the full abstract ▾