Papers

Mobile robots and vehicles carry synchronized multi-camera rigs, yet many streaming 3D foundation models are designed for monocular input, leaving efficient use of rig geometry a challenge. We present StreamRig, a freeze-and-stream framework that builds causal streaming odometry for calibrated rigs on a frozen multi-view 3D foundation model. The frozen front-end jointly perceives the synchronized views using rig calibration. A Rig-Resampler compresses their features, a CausalBridge applies causal attention with a key-value cache, and a lightweight head regresses rig poses. A periodic re-anchoring protocol supports stable pose estimation over long sequences. Only these modules are trained, 74.6M parameters in total, with relative poses as the sole supervision. Our two-stage training strategy combines group relocalization pretraining with causal rig training to transfer the geometric priors of the frozen front-end and the alignment ability of the pretrained modules to streaming odometry. We evaluate on NCLT, TartanGround, KITTI-360, and our self-collected humanoid-robot dataset ZJH, where training uses only simulation and real-world evaluation is zero-shot. Across all four datasets, StreamRig achieves lower translation and rotation drift than the evaluated non-oracle monocular streaming and rig-aware offline models, while maintaining low inference cost. Ablations and controlled camera-count experiments identify the sources of these gains. We further examine how longer training windows affect inference over longer horizons. Code has been released at https://github.com/WeiYuFei0217/StreamRig.

Large language models (LLMs) have shown promise for recommendation reranking, but their use introduces an important trade-off between recommendation quality and serving efficiency. We investigate whether decision-oriented systems provide a useful alternative when the reranking task is fundamentally a structured choice among predefined candidate items. Specifically, we conduct a controlled empirical study of two decision-oriented systems, TypeSafe Jev and OpenAI Decisions, for recommendation reranking. Across three Amazon Reviews domains and four candidate-set sizes, we compare these systems against recommendation-specific models and pointwise and listwise LLM-based rerankers, systematically evaluating recommendation effectiveness, candidate-order robustness, and efficiency and scalability. Our results reveal that Jev frequently achieves a favorable quality–latency trade-off, offering competitive recommendation effectiveness at substantially lower observed latency than stronger LLM-based alternatives. Overall, our findings suggest that specialization for decision-making does not inherently guarantee superior recommendation quality or robustness. Instead, the practical value of decision-oriented inference depends on implementation-specific characteristics and application requirements. These findings motivate further investigation of decision-oriented systems for recommendation and other ranking tasks with structured output spaces.

Converting free-text radiology reports into structured labels supports cohort building, quality assurance, and monitoring of clinical imaging models, but the strongest label extractors are hosted proprietary models whose use raises privacy, cost, and reproducibility concerns. We asked whether a fine-tuned open-weight model (Gemma-3-12B) can match GPT-4o at multi-label intracranial hemorrhage (ICH) acuity extraction from non-contrast head-CT reports, and which ingredients matter. Using a 2x2 design, we crossed two adaptation strategies (a discriminative classification head, CH; generative instruction fine-tuning, IFT) with two training-data sources (distillation of real GPT-4o-labeled reports; synthetic reports generated by GPT-4o from real exemplars), across five training sizes, benchmarked on 100 expert-adjudicated reports against GPT-4o and the un-tuned open-weight base. The distilled instruction-tuned model (DIFT) matched GPT-4o (macro-F1 0.845 vs 0.850; p = 1.000) and exceeded the base model by 0.178. The decisive factor was the training-data source, not the fine-tuning method: both synthetic-data models failed to exceed the un-tuned open-weight base at any training size and underperformed the distilled models across all acuity classes. Fine-tuning and inference fit within the memory envelope of a single 24 GB consumer GPU. For narrow, high-value clinical label-extraction tasks, distilling real reports, rather than generating synthetic ones, is what closes the gap to a hosted model, enabling a private, low-cost, version-stable on-premises alternative.

Continuous diffusion language models generate all tokens in parallel, yet high-quality generation can still require hundreds of network evaluations (NFEs). We study how distributional distillation can reduce this cost by exploiting the student's probabilistic token outputs. Our unified formulation connects the student's output parameterization to the resulting gradient estimators and yields two methods with the same student architecture and reverse-KL matching objective: Simplex-DMD uses continuous token relaxations and pathwise gradients, while Reinforce-DMD uses categorical sampling and REINFORCE with a learned density ratio. We develop both methods for multi-step generation and investigate the training and sampling choices associated with each parameterization. On OpenWebText, for sequences of 1,024 tokens, Simplex-DMD achieves a generative perplexity of 45.6 at a unigram entropy of 5.44 nats in just 4 NFEs, a 49% reduction relative to the strongest evaluated diffusion baseline at matched entropy and sampling budget. Reinforce-DMD improves the frontier at larger budgets, reaching a generative perplexity of 14.9 at an entropy of 5.00 nats with 256 NFEs, a 20% reduction under the same comparison protocol.

Visual perception is conventionally formulated as a one-shot prediction from a single glance at the image, under the assumption that the image content and the model's parametric knowledge suffice to resolve the query. This assumption often fails in real-world scenarios that hinge on fine-grained visual details or require knowledge-intensive and up-to-date information. We term such cases perception under insufficient evidence and formulate perception as an agentic process that can obtain information beyond a single glance. To address the absence of data for this setting, we design two dedicated data generation pipelines, yielding EviRover-SFT-5K and EviRover-RL-12K for training. We further construct EviLens, a human-verified benchmark comprising 688 instances across five perception categories. Building on these data, we present EviRover, to our knowledge the first perception agent explicitly trained to resolve perceptual queries through interaction, using supervised fine-tuning followed by agentic reinforcement learning. Experiments show that the 4B EviRover outperforms its backbone by 30 points on average on EviLens, reaching performance comparable to advanced proprietary models. The gains transfer beyond EviLens to WebEyes, conventional perception benchmarks, and general multimodal benchmarks, including a 15-point improvement on BrowseComp-VL. All code, models, and data are released.

Bosonic quantum systems are among the leading architectures for quantum information processing, offering continuous-variable degrees of freedom with strong error-correction capabilities. However, standard bosonic quantum computation models such as the Lloyd-Braunstein [Lloyd and Braunstein, 1999] and hybrid oscillator-qubit models [Brenner, Dias, and Koenig, 2025; Liu et al., 2026] permit dramatic energy growth, leading to unphysical computational power and the breakdown of fundamental algorithmic tools such as universal and efficient compilation [Brenner et al., 2026; Rudolph et al., 2025]. To address this, we introduce a new model of bosonic quantum computation, Bosonic Energy-Preserving Quantum Computation (BEQC), in which energy is treated as a computational resource. Namely, energy is supplied solely through input coherent states, and all gates are generated by energy-preserving Hamiltonians. Thus, by construction, dramatic energy growth is impossible, making the model physically grounded. We next show that BEQC is a computationally robust and universal model in many respects, including: (1. Computational power) BEQC efficiently simulates all polynomial-energy computations in existing models, and exactly recovers BQP in the polynomial energy setting. It further admits several complexity-theoretic upper bounds when varying the energy, precision, and space parameters of the model. (2. Universal gate sets and state synthesis) BEQC has natural universal gate sets based on linear optics and Kerr interactions. In particular, we obtain a Solovay-Kitaev theorem which circumvents previous no-go results. We give various applications, including (a) a protocol for engineering GKP states with rigorous preparation guarantees, (b) Fock state preparation to exponential precision, and (c) native Fock space simulation of any qubit-based unitary.

Electric ride-pooling must jointly manage passenger service, routes, batteries, and finite chargers; a locally useful relocation can reduce later service. We introduce a discrete-event ride-pooling simulator with ordered passenger stops, pickup and ride-time constraints, battery reserves, charger travel, and finite first-come, first-served (FCFS) charging queues. On top of a common dispatcher, we add a learned repair layer that moves uncollected passengers between vehicles, uses a supervised score to estimate later effects, and uses a two-stage proximal-policy-optimization (PPO) policy to focus computation on promising moves. Every executed move must satisfy route, battery, and time constraints and provide an immediate route-cost improvement. For larger instances, we limit the number of candidate moves and apply several non-conflicting repairs at once. On 12 held-out Manhattan episodes with 100 requests, 30 EVs, and one plug per station, supervised repair reduces operational cost by 6.57% versus no repair (12/12 wins) and serves 74.92 rather than 72.75 requests. Two-stage PPO stays within 1.43% of supervised repair with 95.4% fewer exact trials, while our experiments scale to 8,000 requests and 2,400 EVs with cost reductions across all three larger development settings. Queue-aware charging reduces waiting by 41.0%; under a declared 1.5-kW queue-idle load, operational electricity per served request falls 1.73%.

When a camera revisits a previously observed region, a video world model should reproduce what was there before. This requires both remembering past observations and retrieving the right one for the current viewpoint. Key-value caches preserve visual detail but grow with video length; recurrent memory is compact but compresses history into a fixed-size state, so individual past observations are no longer directly accessible. We introduce LOCI, a hybrid spatial-memory architecture that keeps both representations. In half of the transformer blocks, main attention keeps a key-value cache of past observations; in the other half, it is restricted to the current chunk and complemented by a recurrent linear-attention memory whose reads and writes are conditioned on projective camera geometry, so viewpoint enters both memory addressing and stored content. Recurrent readouts flow into subsequent cache-backed blocks and supply their queries with accumulated scene context. On the public MIND memory benchmark and on held-out recorded trajectories, LOCI reproduces revisited content more faithfully than representative world models and a same-recipe full-softmax model; with full history, it lowers peak memory at equal length by about 30% relative to full softmax. With a bounded bank of retained observations, it streams long videos at constant memory and remains more faithful than full softmax under the same budget.

Training LLM agents with reinforcement learning (RL) is bottlenecked by environments, which must provide verifiable rewards, support long-horizon interaction, and scale cheaply. Existing approaches rely on costly human-curated data or on LLM-generated environments that risk hallucinations and benchmark contamination. We show that LLMs can instead be trained into capable search agents using synthetic environments generated entirely by rules, whose generation requires no LLM and has zero marginal cost. We build PhantomEnvironments, multi-turn RL environments from fictional worlds, where agents must search a corpus of templated articles to answer multi-hop questions. Despite sharing no facts with the real world, these strikingly simple environments yield agents that transfer to real-world multi-hop search benchmarks, often outperforming real-world training data on newer benchmarks. Trained agents generalize to unseen fictional universes, and Qwen models learn to scale their search budget roughly linearly with question difficulty, suggesting emergent search scaling from environment interaction alone. Ablating environment complexity reveals that hop count drives transfer more than constraints or comparisons: even the simplest rule-generated environments are a surprisingly effective, free resource for training generalizable LLM agents.

World Action Models (WAMs) couple visual dynamics prediction with action generation, yet they do not explicitly support the reuse of action experience across manipulation tasks. Furthermore, existing WAMs struggle to capture underlying cross-task semantic relationships that could guide target action prediction, as redundant background elements interfere with the extraction of key visual information. To address these challenges, we develop a novel Action Experience Dictionary (AED) that encodes historical physical action trajectories into shared action embeddings to support skill reuse and model cross-task relationships. Specifically, we first aggregate historical actions to align with visual observations and retrieve action embeddings from the AED using a pretrained action tokenizer. Subsequently, we visually condition the pooled embeddings through cross-attention and prepend them to noisy action tokens, providing interaction context and action intent for prediction. To model action-related motion and reduce reliance on irrelevant background cues, we introduce a motion-aware transition loss that supervises visual feature change prediction over random temporal intervals. Experiments on simulation benchmarks and in real-world cross-embodiment settings verify the effectiveness of our AED. The project code is available at https://github.com/JiahuaDong/AED .

The volume estimation problem is a classic task in computational geometry. The development of randomized algorithms for this problem spurred the development of many influential algorithmic techniques related to Markov Chain Monte Carlo and simulated annealing, and the problem connects to several important geometrical results, like the KLS conjecture. In this work, we quantize the state-of-the-art $\widetilde{O}(d^{3.5}+d^3/\varepsilon^2)$-query randomized algorithm developed by Cousins and Vempala, and obtain a $\widetilde{O}(d^{3.5} + d^{1.75}/\varepsilon)$-query quantum algorithm, improving over the $\widetilde{O}(d^{3.5} + d^{2.25}/\varepsilon)$ state-of-the-art bound. Our key technical contribution is a framework for amortizing the cost of a quantum walk. The framework is based on the recent transducer toolkit introduced by Belovs, Jeffery and Yolcu. It is this amortized quantum walk framework that allows us to exploit the amortized analysis of the ball walk by Cousins and Vempala, thus overcoming the key barrier that previously barred its quantum implementation.

This study compared classifiers that differentiate between urbanized and non-urbanized areas based on unmanned aerial vehicle (UAV)-acquired RGB imagery. The tested solutions in-cluded numerous vegetation indices (VIs) thresholding and neural networks (NNs). The analysis was conducted for two study areas for which surveys were carried out using different UAVs and cameras. The ground sampling distances for the study areas were 10 mm and 15 mm, respectively. Reference classification was performed manually, obtaining approximately 24 million classified pix-els for the first area and approximately 3.8 million for the second. This research study included an analysis of the impact of the season on the threshold values for the tested VIs and the impact of image patch size provided as inputs for the NNs on classification accuracy. The results of the con-ducted research study indicate a higher classification accuracy using NNs (about 96%) compared with the best of the tested VIs, i.e., Excess Blue (about 87%). Due to the highly imbalanced nature of the used datasets (non-urbanized areas constitute approximately 87% of the total datasets), the Mat-thews correlation coefficient was also used to assess the correctness of the classification. The analysis based on statistical measures was supplemented with a qualitative assessment of the classification results, which allowed the identification of the most important sources of differences in classification between VIs thresholding and NNs.

We develop a submodular framework for the von Neumann entropy of discrete quantum convolutions, providing a noncommutative counterpart to the direct side of entropic additive combinatorics. We first introduce globally weighted quantum convolutions, which form compatible families indexed by admissible subsets of a fixed collection of inputs. Our main theorem reveals a polymatroidal geometry underlying their entropy growth: relative to any fixed admissible input block, the entropy gains admit a normalized, monotone, submodular extension to all subsets of the remaining inputs. The theorem yields convolutional strong subadditivity, quantum Ruzsa triangle inequality, and quantum entropic Plünnecke–Ruzsa inequalities for arbitrary input states. For repeated inputs, it gives sharp comparisons of entropy growth across admissible scales; in particular, the quantum doubling constant controls all higher admissible convolution entropies with optimal exponents. Together, these results bring submodular methods from additive combinatorics into the quantum setting and provide a systematic route to broad families of convolutional entropy inequalities.

Extending single Unmanned Aerial Vehicles (UAVs) exploration methods to multi-UAV teams can improve coverage speed and robustness, but introduces challenges such as consistent mapping, safe navigation, and deployment strategy. In this work, we present a centralized multi-UAV exploration framework that enables the use of existing single-UAV sampling-based planners in a multi-UAV setting. The proposed architecture allows multiple UAVs to collaboratively explore unknown environments using a shared global Truncated Signed Distance Field (TSDF) map and centralized planning. Building on the voxblox library, we adapt its mapping pipeline to support real-time fusion of depth measurements from multiple UAVs into a common TSDF representation. In addition, inter-UAV collision avoidance and robot self-filtering mechanisms are integrated into the system to ensure safe navigation and prevent reconstruction of other UAVs as static obstacles. The framework is evaluated in simulation using four sampling-based exploration planners - RH-NBVP, KRH-NBVP, AEP, and KAEP - whose core sampling logic is preserved, with only system-level adaptations for multi-UAV operation. Experiments are conducted across multiple environments and under two deployment configurations: Joint Start (JS), where UAVs are initialized in close proximity, and Separated Start (SS), where UAVs are initialized in distinct locations. Results show that SS deployments consistently achieve faster exploration and improved coverage across all planners, highlighting the importance of the deployment strategy in multi-UAV exploration performance.

Speech-to-speech systems must solve tasks whose requirements emerge across conversational turns. We introduce SpeechConversationBench (SCB), a focused evaluation of spoken mathematical reasoning using 103 sharded GSM8K problems. The framework compares the original problem delivered in one turn (full), its concatenated information shards delivered together (concat), and incremental spoken disclosure across turns (sharded). We report final-answer accuracy for four commercial speech systems and LEGO, a proprietary speech pipeline developed internally by the SCBX Innovation Lab team with explicit conversational context management. Relative to concat, sharded accuracy decreases by 5.0-25.3 percentage points across the four commercial systems. LEGO achieves 77.5 percent accuracy in all three conditions, compared with 76.6 percent sharded accuracy for GPT-4o Realtime. The two single-turn baselines distinguish sensitivity to problem reformulation from the additional challenges introduced by incremental spoken interaction.

Long-term egocentric video enables personalized AI assistants to reason about daily life. However, as video histories grow to hundreds of hours spanning months or years, reprocessing raw clips for every query becomes computationally prohibitive. Memory systems offer a scalable alternative by compacting videos into text representations, but often fail on practical benchmarks: either the memory does not preserve key evidence, or the retriever fails to locate relevant entries due to retrieval competition in growing search spaces. To address these challenges, we introduce MemLife, a multimodal memory system that constructs entity-grounded, first-person text episodes and retrieves them via a time-indexed agentic reader. Without training or query-time video access, MemLife improves over the strongest training-free baseline by 4.6–12.0% across four long-horizon benchmarks. To further improve memory quality, we propose MemOpt, a reinforcement learning framework that optimizes the memory writer to produce faithful, informative, and retrievable memories. MemOpt consistently improves MemLife by 2.7–5.0% across different video and question distributions, with gains that generalize across writer and reader backbones and memory systems.

Quantum one-time programs (Broadbent, Gutoski and Stebila, CRYPTO 2013) or OTPs for short, enable a functionality to be encoded into a quantum token that can be evaluated on a single chosen input and then becomes unusable. While powerful, this primitive is inherently stateless and tied to a setting in which a quantum token needs to be issued and distributed for every single evaluation of a circuit. This raises a natural question: can the one-time computation paradigm be extended to richer, stateful forms of controlled access, and would such an extension offer inherent advantages beyond standard OTPs? We introduce $\textit{query-limited RAM programs}$ (QLPs), a RAM-generalization of QOTPs that supports structured, stateful computation under bounded or policy-driven access. QLPs allow controlled sequences of evaluations while preventing adversarial forking or rollback of computational state. This enables new applications beyond stateless one-time programs, including quantum tokens for Turing Machines whose size depends only on $\textit{code length}$ (and not runtime), transferable $k$-time or budget-limited programs, and low-communication mechanisms for delegating computation in settings such as Software-as-a-Service. To construct QLPs, we introduce $\textit{one-shot programs}$, unifying one-shot signatures (Amos, Georgiou, Kiayias and Zhandry, STOC 2020) with the single effective query paradigm (Gupte, Liu, Raizes, Roberts and Vaikuntanathan, STOC 2025). We prove that one-shot programs generically imply query-limited programs, demonstrating that the strengthened unclonability guarantees of one-shot signatures translate into enhanced functionality. Along the way, we clarify the relationship between signature-token primitives and quantum one-time programs via generic constructions, essentially showing that one-time signing programs imply one-time general computation.

We establish near-linear accuracy bounds for the classical Moreau–Yosida unadjusted Langevin algorithm (MYULA). The target is $\pi\propto e^{-f-g}$, where $f\in C^2(\mathbb{R}^d)$ is $m$-strongly convex with Lipschitz gradient and $g$ is convex and globally Lipschitz. Under an explicit parameter-dependent step-size condition, we bound the invariant-measure bias relative to the Moreau-smoothed target by $\widetilde O(h)$, with only logarithmic dependence on the inverse smoothing parameter in the error coefficient. Combining this estimate with the Moreau approximation bias and Wasserstein contraction gives $\widetilde O(\varepsilon^{-1})$ iterations to make the $N$th-iterate law $\mu_N$ satisfy $\sqrt m\,W_2(\mu_N,\pi)\le\varepsilon$, for fixed model parameters and initialization. We bound the stationary error directly, without assuming third derivatives or a Lipschitz Hessian. Each iteration uses one gradient evaluation and one exact proximal evaluation. The key idea in our analysis is to convert a second-order stationary residual into a Wasserstein bound using a Poisson-based estimate.

Sampling several answers and keeping the one a verifier scores highest is one of the simplest ways to buy accuracy at test time. Its effect is reported as a scaling curve: accuracy against the number $k$ of sampled answers. The curve is cheap to draw and expensive to trust. A budget read off it is chosen after looking at every point, so only a band that covers all budgets at once protects the choice, and on a 100-question benchmark a fixed exact-binomial design needs 192,000 generated answers to certify 64 budgets to within $\pm1/32$ at 95%. Most of that cost pays for the wrong uncertainty. A benchmark is a fixed list of questions; at budget 64, about three quarters of the variance of a selected answer's correctness lies between questions, and an audit that revisits every question need not pay for it. We derive the minimax cost of certifying the whole curve, up to logarithmic factors. It has three parts: calibrating the tail of the score distribution, telling the questions apart, and within-question noise summed along the curve. At a single benchmark the last part sharpens to the variance of one answer's influence under the best allocation of answers to questions, which every valid audit pays and an audit that learns the allocation attains, up to a logarithm, as the precision grows. A paired audit built on an exponential inequality for two independent draws at the same question needs no pilot. On 185 held-out score pools it uses 0.74 times the answers of the cheapest competing certified audit at 64 budgets and 0.53 times at 1,024, and on a newly generated MMLU-Pro study it certified the curve with 79,133 answers, within 0.6% of what a cost law fitted beforehand predicted from the study's within-question variance. The same paths certify pass@$k$ and majority voting, and the bands extend to populations of questions and to answers that depend on earlier ones.

Constrained generation aims to sample from language models (LMs) conditioned on hard constraints. Existing constrained-generation techniques for nondeterministic finite automaton (NFA) constraints either distort the distribution or sacrifice efficiency. Theoretically, this task reduces to counting the length-$n$ sequences accepted by an NFA (#NFA), and the exact #NFA problem is #P-complete. Recent work has shown that #NFA admits a fully polynomial randomized approximation scheme (FPRAS). Inspired by this result, we propose NFA-LM, a polynomial-time engine for NFA-constrained generation with theoretical guarantees under mild assumptions. Experiments show that NFA-LM efficiently generates high-quality outputs with theoretically bounded approximation error.

We develop a mixing time method for estimating the sample complexity of quantum state discrimination. We start with considering the minimum-error discrimination of geometrically uniform pure state ensembles, and prove that its sample complexity has a tight estimate given by a quantum homogeneous mixing time [George et al., 2026] and a quantum version of the generalized Dobrushin coefficient [Wolfer, 2020]. This quantum mixing time further reduces to a classical one when the generating group $G$ forms a Gelfand pair with the stabilizer subgroup $H$ of the generator state. In this case the generalized Dobrushin coefficient can be fully expressed by representation-theoretic quantities of the commutative Hecke algebra $\operatorname{End}_G(\mathbb C[G/H])$. In particular, this method reduces the sample complexity estimation of learning quantum coupon collector states [Arunachalam et al., 2020] and learning phase states to classical mixing time problems. We apply this framework to answer the open problems of learning degree-$d$ phase states over $\mathbb F_q$ in [Alrabiah et al., 2026] and generalized Boolean phase states over $\mathbb Z_q$ [Arunachalam et al., 2023]. The framework also applies to hypergraph state ensembles, giving estimates expressed fully in terms of hypergraph data and recovering estimates for graph state ensembles in [Montanaro and Shao, 2022]. Finally, we extend the discussion to arbitrary mixed state ensembles with uniform priors, prove a sandwiched bound for minimum-error discrimination sample complexity by a quantum weakly mixing time, and provide a tight estimate for the minimax discrimination sample complexity from [D'Ariano et al., 2005] by a Dobrushin-type coefficient. We also discuss the method of strengthened data processing inequality [Gao and Rouzé, 2022] and give an upper bound in terms of a strengthened data processing inequality constant.

We introduce Index-Translate, a multilingual translation model family that combines a shared multilingual foundation with specialized training for general translation, instruction following, speech translation, controlled dubbing, and long-document translation. It includes three model sizes, 2B, 9B, and 35B-A3B, and supports translation in 150 languages, with multilingual instruction following. Evaluations on general translation and complex translation instructions show that Index-Translate outperforms translation models of comparable size and achieves performance comparable to 100B-scale translation models and frontier models. Image-text evaluation shows that translation remains possible without image examples in Index-Translate-specific training, although results vary by model size. Index-Echo provides end-to-end speech-to-text and speech-to-speech translation, outperforming existing end-to-end models and achieving performance comparable to frontier omni models. Index-Homura extends the family to syllable-controlled dubbing. Index-NativeLong introduces native long-document translation with a dedicated task formulation and benchmark. These capabilities support diverse translation tasks, including multilingual content production.

We prove the first non-trivial upper bound for the quantum query complexity of Boolean Matrix Product Verification ($\mathsf{BMPV}$), answering a longstanding open question in quantum query complexity. For $n\times n$ matrices, our upper bound is $\widetilde O(n^{17/12})$, improving on the standard $O(n^{3/2})$ bound obtained using Grover search by Buhrman and Špalek [SODA 2006]. We complement this result by showing an $\Omega(n^{5/4})$ lower bound, which improves over the previous best known lower bound of $\widetilde\Omega(n^{19/18})$ by Childs, Kimmel, and Kothari [ESA 2012]. Our approach centers on a connection with Orthogonal Vectors ($\mathsf{OV}$), which asks whether an indexed list of $n$ Boolean vectors of dimension $n$ contains two vectors with disjoint supports. In particular, we prove equivalences between $\mathsf{OV}$ and $\mathsf{BMPV}$ and establish the above bounds for $\mathsf{OV}$. We also prove a tight $\widetilde \Theta(n^{3/2})$ bound for a variant of $\mathsf{BMPV}$ that asks whether the product contains a given row vector. Together, these results imply a polynomial separation between the quantum query complexities of deciding whether a graph has radius at most two and whether it has diameter at most two.

Open concrete canals play a vital role in water transportation, serving as primary water infrastructure for millions of people across the Phoenix, Arizona, metro area. Over time, the concrete canals can experience a range of issues, including canal lining deformation, cracked concrete, and sediment buildup on the canal floor. Identifying such critical issues is a resource-intensive process, which currently happens only during four-year dry-up cycles. This prevents the maintenance crew from prioritizing operations on the most affected canal segments. To address this issue, the research team has developed and verified an easily deployable and non-invasive method to inspect canal beds without draining the water. This inspection system integrates affordable, off-the-shelf drone and sonar technology (termed dronar). This dronar system includes a consumer-grade sonar system integrated into an unmanned surface vehicle (USV) that carries the sonar transducer just under the surface of the canal water. This paper presents a proof-of-concept demonstration of the dronar system across three field tests on the Arizona Canal in Phoenix. DownScan depth profiles from the sedimented canal segment were consistently shallower than profiles from the same segment after cleaning, with offsets of up to 15 cm observed along the track. Repeated runs over the clean segment produced closely overlapping DownScan depth profiles, confirming that the dronar yields repeatable measurements across the natural variation of the canal bed. These results establish the dronar as a viable proof-of-concept tool for non-invasive canal bed inspection.

Safe social navigation requires a robot to anticipate not only the future consequences of its actions, but also whether a nominal action can actually be executed under surrounding physical and social constraints. We present Social-WM, an efficient latent world-model planning framework trained from egocentric RGB video sequences. Our key observation is that social-navigation experience contains a systematic discrepancy between the nominal action and the realizable action: a nominal forward action may be fully executed in free space, but needs to be constrained when heading towards a pedestrian or obstacle. Social-WM learns these safety-relevant consequences directly through action-conditioned future prediction, where the target is the actual observed future following each command. We further introduce a realizable inverse-dynamics objective that associates observed latent transitions with the action actually realized rather than the nominal one. At deployment, candidate actions are imagined through the latent world model, and the inverse dynamics model estimates their realizability; nominal–realizable discrepancy then provides a safety signal before execution. The learned dynamics and realizability model remain goal-independent and support both position- and image-goal navigation. On Social-HM3D, Social-WM achieves 63.77% success while reducing human collisions to 21.67%, and maintains strong performance under zero-shot transfer to Social-MP3D, without explicit pedestrian tracking, privileged human state, or online reinforcement learning.

MANETs enable flexible infrastructure-less wireless connectivity in dynamic and resource-constrained environments. As modern MANETs exploit multiple frequency channels and support heterogeneous traffic patterns, decentralized transmit-power allocation becomes increasingly challenging. We develop a unified learned optimization framework for decentralized power allocation in dynamic multi-hop, multi-channel MANETs. We formulate a constrained end-to-end throughput maximization problem covering unicast, multicast, multicommodity, convergecast, and many-to-many communication. Although centralized and non-convex, this problem serves as an unsupervised training objective for MANET-GNN, a message-passing GNN that operates as a distributed learned optimizer. MANET-GNN uses only local, possibly noisy, CSI and a prescribed number of neighbor message exchanges, enabling low-latency decentralized inference while generalizing across topologies and network sizes. Numerical results show that MANET-GNN achieves centralized-competitive performance across communication frameworks, remains robust to channel uncertainty, and scales effectively across MANET configurations.

Language agents are expected to solve increasingly complex tasks, creating a growing need for continual improvement. One promising approach is to evolve the agent harness, the software that governs tool use, memory management, and task execution, while keeping the underlying language model fixed. Recent methods automate this process by using a meta coding agent to modify the harness based on execution feedback. However, relying on that agent's existing knowledge and observed failures can restrict exploration and make adaptation reactive. Inspired by how human experts learn from the research literature for new solutions, we introduce ScholarEvolve, a framework that automatically draws on state-of-the-art research to guide harness evolution. ScholarEvolve organizes the harness evolution directions into functional modules and uses topic modeling to identify distinct improvement strategies for each module. It implements these strategies and evaluates their combinations to improve task performance. Moreover, the framework is designed to incorporate new publications over time, allowing research advances to drive proactive lifelong evolution. Experiments demonstrate improvements on AppWorld and Tau2-Bench. ScholarEvolve raises Qwen3.5-27B task goal completion from 49.6% to 63.6% on AppWorld Challenge, and raises GPT-5.4-mini pass@1 from 72.7% to 81.9% on Tau2-Bench Telecom.

We introduce shadow quantum singular value transformation (Shadow QSVT): given an initial state $|{\psi}\rangle$, a Hermitian matrix $H$, a polynomial $f$, and a set of observables $\{O_1,\dots,O_m\}$, the goal is to estimate $\langle{\psi}|f(H)^{\dagger}O_j f(H)|{\psi}\rangle$ for all $j\in\{1,\dots,m\}$. Shadow QSVT provides a systematic route to reduce the quantum resources required by standard QSVT, which constructs a unitary block-encoding of $f(H)$. It uses structure in the input state and observables, together with the fact that many applications require only observable estimates rather than synthesizing the full unitary. We present three algorithms that exploit structure in the initial state and observables to reduce quantum circuit depth. First, we develop a state-aware QSVT algorithm that prepares the target state with low circuit depth when the Krylov subspace associated with $H$ and $|\psi\rangle$ is low-dimensional or admits an accurate low-dimensional approximation. Second, we introduce an observable-aware Shadow QSVT algorithm that combines a new observable-aware Krylov subspace with history states to further reduce circuit depth and gate complexity. Finally, we develop Classical Shadow QSVT, which constructs a classical representation from $H$, $f$, and $|\psi\rangle$ without prior knowledge of the observables or explicit preparation of the target state proportional to $f(H)|\psi\rangle$. This representation enables estimation of the target quantities for observables specified after the quantum computation. Together, these three algorithms provide tools for reducing the circuit depth of QSVT-based computations across a range of settings.

We present PrefPI (Preference-Guided Policy Iteration), an iterative framework for steering pretrained generative robot policies using only relative preferences over self-generated trajectories. Unlike prior preference-learning methods that primarily sharpen modes already represented by the policy, we study steering beyond the initial effective support, where desired behaviors are rarely or never observed under the initial policy. Our key idea is to formulate preference learning as preference-conditioned generative modeling: preferred trajectories define a conditional distribution, whose density ratio with the broader behavior prior provides an implicit preference signal amplified by classifier-free guidance (CFG). Repeating this preference-conditioned modeling and guidance step yields a form of preference-guided policy iteration, turning incremental improvements toward previously inaccessible behaviors. Across diffusion policies and the PI0.5 flow- matching VLA in simulation and the real world, PrefPI produces substantial behavioral shifts with limited feedback. In particular, PrefPI increases object transport height from 10.7 cm to 19.8 cm on real hardware with only 150 preference-labeled trajectories.

Sparse triangular solve (SpTRSV) is a fundamental kernel in numerous scientific and engineering applications. However, the data dependencies inherent in sparse triangular matrices significantly limit the available parallelism and make efficient workload distribution challenging. Recent graph transformation techniques address these limitations by modifying the dependency graph of the input matrix to improve parallel execution. Existing graph transformation strategies, however, rely on manually designed heuristics, making their development and adaptation to different optimization objectives challenging. This work proposes a reinforcement learning-guided graph transformation framework for SpTRSV, in which graph transformation is formulated as a sequential decision-making problem and an RL agent learns matrix-dependent transformation policies. Experimental results on real-world sparse matrices demonstrate level reductions of up to 94% and reductions of up to 80% in the coefficient of variation of level costs, while modifying only 1.50% of the rows in the highest case. On average, the RL- guided graph transformation achieves a 23% reduction in the number of levels and a 29% reduction in the coefficient of variation of level costs while rewriting only 0.82% of the matrix rows. Although the heuristic strategies generally achieve more aggressive level reduction(between 31% and 46%), the RL-based approach achieves the largest average reduction in the coefficient of variation of level costs, demonstrating its ability to balance competing graph transformation objectives. The results further show that the learned policies can be transferred to previously unseen matrices through curriculum learning and fine-tuning, while zero-shot experiments provide insights into the limitations of generalizing graph transformation policies across different sparsity patterns.