arXiv:2606.19402v1 Announce Type: new Abstract: We investigate the evolution of cosmic structures in the power-law modified gravity model $f(R)=R+R^{1+\delta}/R_c^\delta$, where the dimensionless parameter $\delta$ characterizes deviations from General Relativity. The background cosmological dynamics and the evolution of linear matter density perturbations are studied within the framework of metric $f(R)$ gravity. The modified perturbation equation is derived by introducing an effective gravitational coupling associated with the additional scalar degree of freedom, and the evolution of the growth factor, logarithmic growth rate, growth index, and the observable quantity $f\sigma_8(z)$ are investigated. The results show that the curvature correction enhances the growth of matter perturbations while remaining compatible with the observed late-time accelerated expansion for suitable values of the model parameter. The theoretical viability of the model is established through the ghost-free condition, Dolgov--Kawasaki stability criterion, positive scalaron mass, stable de Sitter solution, and chameleon screening mechanism. Comparison with representative viable $f(R)$ gravity models shows that the present theory achieves a consistent cosmological evolution with a single deviation parameter. The predicted modifications in the growth of structures and the effective gravitational coupling provide observable signatures that can be tested by forthcoming large-scale structure and weak-lensing surveys, providing a means to test curvature-induced modifications of gravity.
Science Journals
arXiv:2606.19458v1 Announce Type: new Abstract: We present MonaVec, a deterministic, embedded vector-search kernel for edge and offline AI -- settings where server infrastructure, network connectivity, and training data are all unavailable. Existing vector-search systems assume a persistent server, gigabytes of RAM, or a training pass over the corpus; MonaVec instead targets the deployment profile of SQLite: one file, one function call, runs anywhere. Its quantization core is training-free by default and data-oblivious: a Randomized Hadamard Transform (RHDH) conditions any input distribution toward N(0,1), so precomputed Lloyd-Max tables quantize to 4 bits (8x smaller) with no learned codebook and no data pass. The index persists as a single .mvec file whose embedded ChaCha20 rotation seed makes results reproducible across architectures and byte-identical within a build -- a determinism guarantee that parallel-build graph libraries cannot offer. On semantic embeddings (AG News, 45K x 1024-dim BGE-M3, cosine), MonaVec 4-bit BruteForce reaches 0.960 Recall@10 in 27 MB -- leading float32 FAISS-IVF and 8-bit usearch on recall -- while trading peak throughput for byte-identical determinism. A single-pass global standardization (fit()) extends the same data-oblivious pipeline to magnitude-sensitive L2 data, and optional IvfFlat and HNSW backends carry it to million-vector corpora. MonaVec is implemented in pure Rust with Python bindings and runtime SIMD dispatch (AVX-512/AVX2/NEON/scalar). It targets on-device RAG, offline agents, and embedded retrieval -- the niche SQLite occupies for relational data: one file, one call, runs anywhere.
arXiv:2606.19460v1 Announce Type: new Abstract: We introduce the first generative foundation model for chest radiograph synthesis trained from scratch at the billion-parameter scale. Existing radiographic AI models often suffer from poor generalisation across patient subpopulations, institutions, and acquisition settings, resulting in limited real-world clinical utility. Controlled, high-fidelity synthesis of chest radiographs is a promising path toward diversifying clinical datasets and evaluating the robustness of diagnostic models. Therefore, we present the largest specialist generative foundation model for chest radiographs to date, with over 1.3B parameters, trained for 1.6T tokens on a curated, heterogeneous dataset comprising 1.2M radiographs and clinical expert-guided metadata. Our model supports controllable radiograph generation and editing across multiple demographic subgroups, acquisition views, and a dozen pathologies. Moreover, we significantly advance the state of the art in radiograph synthesis fidelity, producing images that are indistinguishable from real radiographs to clinical experts.
arXiv:2603.09420v3 Announce Type: replace Abstract: Motion forecasting enables autonomous vehicles to anticipate scene evolution by predicting the future trajectories of dynamic agents. However, existing approaches typically assume a closed-world setting with a fixed object taxonomy and access to high-quality perception, limiting their applicability in the real world where perception is imperfect, and new object classes may emerge over time. In this work, we introduce class-incremental motion forecasting, a novel setting in which new object classes are sequentially introduced over time and future object trajectories are predicted directly from camera images. We propose the first end-to-end framework for this setting, which adapts to newly introduced classes while mitigating catastrophic forgetting of previously learned ones. Our method generates motion forecasting pseudo-labels for known classes and matches them with 2D instance masks from an open-vocabulary segmentation model. This 3D-to-2D keypoint voting mechanism filters inconsistent and overconfident predictions, while a query feature variance-based replay strategy samples informative past sequences to preserve prior knowledge. Extensive evaluations on nuScenes and Argoverse 2 show that our approach successfully preserves performance on known classes while effectively adapting to novel ones. We further demonstrate zero-shot transfer to real-world driving and show that the framework extends naturally to open- and closed-loop end-to-end class-incremental planning on nuScenes and NeuroNCAP. Code and models will be made publicly available at https://omen.cs.uni-freiburg.de.
arXiv:2603.15106v2 Announce Type: replace Abstract: Enabling efficient deep neural network (DNN) inference on edge devices with different hardware constraints is a challenging task that typically requires DNN architectures to be specialized for each device separately. To avoid the huge manual effort, one can use neural architecture search (NAS). However, many existing NAS methods are resource-intensive and time-consuming because they require the training of many different DNNs from scratch. Furthermore, they do not take the resource constraints of the target system into account. To address these shortcomings, we propose PrototypeNAS, a zero-shot NAS method to accelerate and automate the selection, compression, and specialization of DNNs to different target microcontroller units (MCUs). We propose a novel three-step search method that decouples DNN design and specialization from DNN training for a given target platform. First, we present a novel search space that not only cuts out smaller DNNs from a single large architecture, but instead combines the structural optimization of multiple architecture types, as well as optimization of their pruning and quantization configurations. Second, we explore the use of an ensemble of zero-shot proxies during optimization instead of a single one. Third, we propose the use of Hypervolume subset selection to distill DNN architectures from the Pareto front of the multi-objective optimization that represent the most meaningful tradeoffs between accuracy and FLOPs. We evaluate the effectiveness of PrototypeNAS on 12 different datasets in three different tasks: image classification, time series classification, and object detection. Our results demonstrate that PrototypeNAS is able to identify DNN models within minutes that are small enough to be deployed on off-the-shelf MCUs and still achieve accuracies comparable to the performance of large DNN models.
arXiv:2606.19523v1 Announce Type: new Abstract: Modern stellarators are typically designed by optimizing the shape of the plasma boundary surface, with the parameters taken to be Fourier amplitudes. Many promising optimization algorithms such as Bayesian methods require bound constraints on the parameters and are most efficient when each parameter is scaled similarly to the others. With the typical Fourier parameterization, it is unclear how to set these bounds: wide constraints lead to self-intersecting boundaries and frequent failures of the MHD equilibrium calculation, while tight bound constraints limit expressiveness. To address these issues, here we propose two new parameter spaces for stellarator optimization. Both begin with a dataset of existing stellarator boundaries. In the first approach, a quantile transformation is applied to each Fourier degree of freedom, mapping the data distribution to a uniform distribution on the unit interval. In the second approach, principal component analysis (PCA) is applied to points on the boundaries, followed by a quantile transformation. For both approaches, the transformed variables become the degrees of freedom, naturally bounded to [0, 1]. The PCA method has the additional benefit of dimensionality reduction, with high expressiveness for a small number of parameters. The methods are demonstrated via Bayesian optimization for good alpha-particle confinement with guiding-center tracing inside the optimization loop, using asynchronous parallelization. These optimizations yield stellarator configurations with excellent fast-particle confinement in fields that can be far from quasisymmetric or quasi-isodynamic.
arXiv:2606.19524v1 Announce Type: new Abstract: We introduce Vistas, a tool for visualizing high-energy particle physics collisions simulated by the Pythia Monte-Carlo event generator. Vistas utilizes the browser-based event display framework Phoenix to show distinct computational stages of a high-energy collision event simulation: the hard process, parton shower, hadronization, and particle decays. Particles produced from each of these stages are represented as lines in an interactive three-dimensional graph structure, where each line is along the direction of its particle's three-momentum vector. The event can be rotated, translated and zoomed, and details for each particle can be accessed by selecting the relevant particle line. Additionally, particle lines from all stages of the simulation can be toggled on and off and can be filtered by particle-level kinematic selection requirements. This interactive environment provides an intuitive interpretation of Pythia simulation output, including detailed features such as color flow, beam remnants, and multiple parton interactions, making it a useful tool in physics education settings, from outreach activities to graduate particle-physics courses.
arXiv:2606.19595v1 Announce Type: new Abstract: Voice agents deployed in structured workflows (customer service, healthcare scheduling, account management) must handle frequent user interruptions while maintaining progress through multi-step procedures. Existing benchmarks for speech-capable models focus on the timing of interruptions: barge-in detection, endpointing, and turn-taking dynamics. They leave unmeasured what happens after the interruption: does the agent resume the workflow at the correct step? Does it address the user's interjection? Does it avoid re-delivering content the user already heard? We introduce IHBench (Interruption Handling Benchmark), a benchmark that evaluates post-interruption recovery in voice agents executing state-machine-driven workflows across 10 enterprise domains. Six interruption types are injected at controlled points mid-utterance, with per-interruption evaluation rubrics generated alongside the data. Each interruption is scored on two axes: task fulfillment and recovery quality. We evaluate 27 audio-language model configurations from OpenAI, Google, and the open-weight community. Models vary widely, and recovery quality depends strongly on the interruption type. Across our experiments, closed-weight models are consistently more robust to interruptions than open-weight ones: they win far more often on task fulfillment, degrade roughly 3.3x more slowly as conversations grow longer, and show no audio-versus-text modality gap, whereas the open-weight models lose ground on all three. A human study validates the LLM judge against human annotators, and a cross-benchmark analysis against AudioMultiChallenge indicates that recovery quality is a largely distinct capability axis.
arXiv:2606.17832v2 Announce Type: replace Abstract: Large language models (LLMs) are often hypothesized to perform implicit Bayesian inference, yet a key coherence condition, the martingale property of predictive beliefs, has been shown to fail in controlled synthetic in-context learning settings. We revisit this question in a more typical usage regime: generic multiple-choice question answering. Exploiting the discrete answer space, we compute exact predictive distributions and study belief dynamics induced by autoregressive answer resampling. We introduce prompted predictive resampling (PPR), where an LLM generates a sequence of answers to the same question. Empirically, PPR reveals early-stage belief drift, indicating martingale violations. However, after sufficient resampling steps, the belief process self-stabilizes and converges to a coherent predictive distribution. Based on this observation, we further propose (i) a seed-answer prompting strategy to accelerate stabilization, and (ii) a self-consistency loss that amortizes early-stage drift into the model via fine-tuning. Experiments on multiple-choice QA benchmarks show that our methods substantially reduce belief drift and improve predictive coherence without sacrificing accuracy.
arXiv:2606.19607v1 Announce Type: new Abstract: Preference-based post-training has become a central paradigm for aligning language models. A common data-collection strategy is to generate a small set of completions for each prompt and label the resulting comparison pairs. However, human preference labels are often much more expensive than generating additional completions, suggesting a different use of the same labeling budget: generate a larger pool of completions, but label only the most informative comparison pairs. This paper studies which pairs should be compared in preference-based post-training. We formulate comparison curation as a sampling-design problem and evaluate designs by the quality of the final policy under the preference-based post-training objective. We instantiate this framework for Direct Preference Optimization (DPO), analyzing how the choice of labeled pairs propagates through DPO training to downstream policy performance. Our main results provide matching upper and lower bounds on the post-training optimality gap of the DPO-trained policy. The bounds show that comparison selection affects downstream performance through a single design-dependent information matrix, which links label allocation to parameter estimation error and policy suboptimality. This yields an explicit optimization criterion for budgeted comparison curation and motivates practical sampling designs for selecting informative pairs from large generated completion pools. Experiments on synthetic settings and language-model post-training benchmarks show that the proposed designs consistently improve sample efficiency over common comparison-selection heuristics.
arXiv:2606.19613v1 Announce Type: new Abstract: We introduce StaminaBench, a benchmark that measures the stamina of coding agents: how many consecutive interaction turns (change requests) they can handle before failing. Unlike the prevailing fraction-of-tasks-solved metric, this matches real vibe-coding where sessions run dozens or hundreds of turns. In StaminaBench, agents implement a REST API server and modify it across a tunable number of procedurally generated follow-up change requests - 100 in our experiments, resulting in codebases of up to 6,000 lines. Tests are generated fully programmatically without LLM involvement, ensuring reproducibility and reliability; change sequences are drawn from either a hardcoded or LLM-driven sampler, both constrained to a structured action space to ensure changes are valid. The agent and the server run in an isolated environment and communicate with the benchmark through HTTP, making testing fully black-box and language-agnostic. We evaluate six agent harnesses paired with seven open-source LLMs across 20 scenarios of 100 turns each and find that: (1) all the tested models fail within 5-6 turns, confirming that vibe-coding-style programming without thorough testing produces bugs; (2) passing test feedback back to the agent and allowing it to retry improves passed turn count by up to 12x; and (3) a good harness is required for strong performance: stronger models exhibit up to a 6x gap between their best and worst harness, while weaker models fail with any harness. We release the benchmark and the generated tasks to enable further research into multi-turn coding agent behavior. Benchmark code and data: github.com/amazon-science/StaminaBench.
arXiv:2606.19620v1 Announce Type: new Abstract: We present G-Lox (group-adaptive Lox), a bridge-distribution system that preserves Lox-style distributor blindness while enabling hidden, stateful group-level adaptation. G-Lox places adaptive assignment logic behind a two-server privacy wall, so no single server learns group identifiers or group-to-bridge assignments. Private state access and state-dependent updates use two-server DPF/FSS protocols and secure two-party computation, supporting blockage reporting, transport-aware reassignment, and privacy-preserving group splitting. We evaluate G-Lox through system measurements and policy simulation. In our C++/EMP implementation over real TCP sockets, private state access has low client-visible overhead: across state sizes up to 2^16, communication remains in the low-KiB range per iteration. At M=1024, the client sends 1,968 bytes, receives 1,280 bytes, and completes an iteration in about 0.25 s. Simulations with group-specific blocking and Sybil enumeration show that G-Lox improves robustness over Lox- and rBridge-like baselines among systems that maintain broad issuance.
arXiv:2402.08727v3 Announce Type: replace-cross Abstract: There has been a surge of recent interest in the Wigner's Friend paradox, sparking several novel thought experiments and no-go theorems. The main narrative has been that Wigner's Friend highlights a counterintuitive feature that is unique to quantum theory, and which is closely related to the quantum measurement problem. Here, we challenge this view. We argue that the gist of the Wigner's Friend paradox can be reproduced without assuming quantum physics, and that it underlies a much broader class of enigmas in the foundations of physics and philosophy. To show this, we first consider several recently proposed Extended Wigner's Friend scenarios, and demonstrate that some of their implications for the absoluteness of observations can be reproduced by classical thought experiments that involve the duplication of agents. Crucially, some of these classical scenarios are technologically much easier to implement than their quantum counterparts. Then, we argue that the essential structural ingredient of all these scenarios is a feature that we call "Restriction A": that a physical theory cannot give us a probabilistic description of the observations of all agents. Finally, we argue that this difficulty is at the core of other puzzles in the foundations of physics and philosophy, and demonstrate this explicitly for cosmology's Boltzmann brain problem. Our analysis suggests that Wigner's Friend should be studied in a larger context, addressing a frontier of human knowledge beyond quantum foundations: to obtain reliable predictions for experiments in which these predictions can be privately but not intersubjectively verified.
arXiv:2509.10416v2 Announce Type: replace Abstract: We present TASC, a Task-Aware Shared Control framework for relational telemanipulation that infers task-level user intent and provides assistance from motion-only input. To support prehensile relational tasks without predefined templates, TASC constructs an open-vocabulary interaction graph from visual input to represent functional object relationships, and infers user intent accordingly. A shared control policy then provides assistance during both grasping and object interaction, guided by spatial constraints predicted by a vision-language model. Our method addresses two key challenges in relational telemanipulation under shared control: (1) task-level intent inference from low-level motion commands, and (2) generalizable assistance across diverse objects and tasks. Experiments in both simulation and the real world demonstrate that TASC improves task efficiency and reduces user input effort compared to prior methods, while enabling zero-shot generalization across diverse relational telemanipulation tasks. The code that supports our experiments is publicly available at https://github.com/fitz0401/tasc.
VCG: A Multimodal Retrieval Framework for E-Commerce Video Feeds under Extreme Cold-Start Conditions
arXiv:2606.19627v1 Announce Type: new Abstract: The digital commerce landscape is shifting from static, search-driven catalogs to dynamic, immersive video feeds. This transition introduces an ``extreme cold-start'' problem: unlike traditional items, new short-form videos lack the dense interaction history required for collaborative filtering. Furthermore, immersive feeds introduce strong position and duration biases that distort standard engagement signals. In this paper, we demonstrate the Video Candidate Generation (VCG) system, a scalable multimodal retrieval engine designed to solve these challenges in a large-scale e-commerce environment. By leveraging a domain-adapted vision-language model (based on CLIP), we map users and videos into a shared semantic space, enabling zero-shot retrieval based on visual content rather than behavioral history. We detail the system's architecture and present a rigorous evaluation comparing generative (LLM) vs. discriminative (CLIP) embeddings. Our results show that while generative models excel at attribute prediction, they suffer from embedding space collapse in retrieval tasks. Online A/B testing demonstrates that VCG effectively mitigates engagement biases, yielding a 50\% uplift in deep video completion. To showcase the system's capabilities, we present an interactive demonstration featuring three bi-directional retrieval scenarios: Product-to-Video, Video-to-Product, and Zero-Shot Semantic Search.
arXiv:2606.19667v1 Announce Type: new Abstract: Retrieval-Augmented Generation (RAG) improves factual grounding, but it also lengthens prompts and raises prefill cost. Prefix caching in serving engines such as vLLM reduces this cost only when requests share the same token prefix. In grounded generation, however, adjacent queries may retrieve overlapping evidence in different orders, so set overlap does not become reusable prefix overlap. We present CacheWeaver, a lightweight prompt-layer method for cache-aware evidence ordering. The method keeps a prefix tree over recently served evidence sequences and uses a greedy walk to place the most reusable prefix first, while leaving the serving engine and retrieved evidence set unchanged. Across three vLLM configurations, the method lowers median time-to-first-token (TTFT) by about 20-33 percent relative to retrieval-order prefix caching, without hurting answer quality in our QA tests. The greedy policy reaches 97.5 percent of the median TTFT gain from oracle ordering, indicating that most reusable prefix locality can be recovered by a simple scheduling layer between retrieval and inference.
arXiv:2606.19350v1 Announce Type: new Abstract: Large language models (LLMs) excel at multi-step reasoning but incur substantial inference cost. We introduce Causal Attribution Pruning (CAP), a training-free method that identifies critical attention heads by measuring their causal impact on reasoning tasks and uses these head-level scores to guide fine-grained weight pruning. For each attention head, CAP estimates the expected performance degradation when the head is masked during forward passes on a small calibration set of reasoning problems. These causal scores are then converted into weight-level importance values for the corresponding projection matrices. Unlike magnitude-only or activation-based criteria, CAP's interventional measurement directly captures each head's functional contribution, yielding relative accuracy gains of up to 61% over Wanda on ARC-Challenge at 20% sparsity. We evaluate CAP on GSM8K, StrategyQA, and ARC-Challenge using Llama-3-8B-Instruct and Mistral-7B-Instruct at 10%, 20%, and 50% sparsity. At moderate sparsity (10-20%), CAP improves over Wanda in most model-benchmark configurations. with especially large gains on ARC-Challenge for Llama-3. Our results suggest that attention-head-level causal attribution can better preserve reasoning performance on downstream benchmarks than correlational pruning criteria at equivalent sparsity, while remaining limited by coarse MLP attribution at 50% sparsity.
arXiv:2606.19668v1 Announce Type: new Abstract: Multilingual Large Language Models (MLLMs) are increasingly expected to handle Code-Switched (CS) inputs, yet mixing languages frequently degrades performance relative to source- or target-language monolingual counterparts. To understand this degradation, we use grammar-forced CS as a controlled diagnostic setting for locating CS representations relative to their source and target counterparts. We introduce Anchor Bias, a geometric measure that quantifies language anchoring, whether a CS hidden state aligns closer to its source or target language counterpart. Across diverse MLLMs, Anchor Bias reveals a consistent grammar-frame effect: source-framed CS stays source-anchored, whereas target-framed CS shifts target-ward and shows larger Question Answering (QA) degradation. Motivated by this representational pattern, we propose CANVAS (Contextual Anchor-based Neural Vector Alignment Steering), an inference-time intervention that extracts a source-side canvas from the input and softly steers target-language hidden states toward the source anchor during prefill. CANVAS consistently recovers QA F1 across MLLMs and CS conditions, showing that internal anchoring signals provide an actionable target for mitigating CS inference failures.
arXiv:2606.19747v1 Announce Type: new Abstract: Quran Automatic Speech Recognition (ASR) aims to convert Quranic recitation into text, enabling applications such as aided memorisation tools and Quranic search engines. However, existing ASR models often exhibit high Word Error Rates (WER) on user-recited verses and lack full coverage of the Quranic corpus. This paper presents a systematic empirical study of domain-specific fine-tuning of pretrained Transformer-based models for Quranic ASR, using advanced speech feature extraction methods: Wav2Vec2.0, HuBERT, and XLS-R. These models apply self-supervised learning by masking portions of input audio and using Transformer architectures to learn context-aware speech features. The pretrained models are fine-tuned on a filtered Quranic dataset exceeding 870 hours of professional and user recitations. Through comprehensive ablation studies across feature extractors, output label formats, training strategies, and clip durations, we identify the key factors that affect transcription accuracy in this domain. Our best-performing configuration achieves a WER of 0.08 on the EveryAyah subset and 0.11 on the combined EveryAyah+Tarteel setting, representing roughly a five-percentage-point gain over the Citrinet baseline (WER = 0.163) while reducing combined-model training time from 140 hours to 40 hours. Arabic text without diacritics yields the best fine-tuning results, and Wav2Vec2-XLSR-53 provides the strongest overall representation. Future work includes improving dataset quality and developing phoneme-aware models to extract deeper speech feature representations for Tajweed-sensitive applications.
arXiv:2512.17473v3 Announce Type: replace-cross Abstract: We present an algorithm based on the alternating direction method of multipliers (ADMM) for solving nonlinear matrix decompositions (NMD). Given an input matrix $X \in \mathbb{R}^{m \times n}$ and a factorization rank $r \ll \min(m, n)$, NMD seeks matrices $W \in \mathbb{R}^{m \times r}$ and $H \in \mathbb{R}^{r \times n}$ such that $X \approx f(WH)$, where $f$ is an element-wise nonlinear function. We evaluate our method on several representative nonlinear models: the rectified linear unit activation $f(x) = \max(0, x)$, suitable for nonnegative sparse data approximation, the component-wise square $f(x) = x^2$, applicable to probabilistic circuit representation, and the MinMax transform $f(x) = \min(b, \max(a, x))$, relevant for recommender systems. The proposed framework flexibly supports diverse loss functions, including least squares, $\ell_1$ norm, and the Kullback-Leibler divergence, and can be readily extended to other nonlinearities and metrics. We illustrate the applicability, efficiency, and adaptability of the approach on real-world datasets, highlighting its potential for a broad range of applications.
arXiv:2604.11774v2 Announce Type: replace-cross Abstract: Neutrons are important final-state particles in neutrino interactions, yet they are not considered or reconstructed in most current neutrino LArTPC physics analyses. In this paper, we present a simulation-based proof-of-concept study of neutron reconstruction in a generic LArTPC detector. Leveraging isolated, MeV-scale energy deposits, or blips, from neutron inelastic scattering, and using realistic blip response from published experimental results, we demonstrate the capability to identify neutrons and to reconstruct the direction and energy of the final-state neutron system in sub-GeV neutrino interactions. We then explore how neutron-related blip attributes can be used to improve physics studies of neutrino interactions, such as enhancing neutrino-antineutrino separation in atmospheric neutrinos and reverse-horn-current beam neutrinos. This simple study provides an initial quantification of LArTPC neutron reconstruction capabilities, which we expect to improve with future advancements in blip reconstruction, identification, and classification algorithms, as well as the modeling of neutrons.
arXiv:2606.19534v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) have achieved remarkable progress in visual understanding tasks. However, most existing MLLMs rely on autoregressive generation, which limits their efficiency for perception tasks that require captioning multiple regions. In this work, we propose PerceptionDLM, a multimodal diffusion language model optimized for efficient parallel region perception. Built upon PerceptionDLM-Base, a strong foundational baseline that achieves state-of-the-art performance among open-source diffusion MLLMs, our architecture fully leverages the parallel decoding nature of DLMs. Specifically, we introduce efficient prompting and structured attention masking to enable simultaneous perception of multiple masked regions, allowing the model to generate region descriptions in parallel at both the sequence and token levels. This design significantly improves inference efficiency compared with existing approaches that process regions sequentially. To systematically evaluate the parallelism property of visual perception capability for DLMs, we construct a new Parallel Detailed Localized Captioning Benchmark (ParaDLC-Bench) by scaling the DLC-Bench to include multiple region masks per image, enabling joint evaluation of both caption quality and inference efficiency. Experiments demonstrate that PerceptionDLM maintains competitive performance in region captioning while achieving substantial speed improvements for multi-region perception tasks. Our results highlight the potential of multimodal diffusion language models for efficient, parallel visual perception. To the best of our knowledge, we are the first to achieve parallel region caption and perception by leveraging the advantages of diffusion language models. Code, models, and datasets are released.
arXiv:2606.19576v1 Announce Type: new Abstract: Remote and disaggregated memory tiers expand the effective memory capacity of analytical database engines, but they also reshape the cost structure of out-of-memory query processing. When an operator spills beyond local DRAM, moving pages to remote memory incurs both data-transfer time and a fixed round-trip latency per transfer. Classical operator analyses and buffer-allocation heuristics primarily target disk spilling by minimizing total I/O volume. Under remote memory, these strategies can be suboptimal because they may trigger excessive transfer rounds. We present REMOP, a remote-memory-aware operator optimization framework that uses transfer-round-aware intra-operator memory policies to improve out-of-memory execution under tight memory budgets. REMOP introduces the number of transfer rounds into the latency cost model and derives operator-specific buffer-partitioning strategies, instantiating the approach for blocked nested-loop join, external merge sort, and external hash join in DuckDB. Our evaluation on a two-node compute-memory testbed shows that REMOP reduces transfer rounds by up to 97% and operator runtime by up to 48% on spill-heavy microbenchmarks, and lowers the average runtime of spilling TPC-H and TPC-DS queries by 22.7% and 26.4% end-to-end.
arXiv:2606.19597v1 Announce Type: new Abstract: Mean opinion scores (MOS) are widely used for speech quality assessment, yet scalar labels are sensitive to rater variability and listening test differences. This introduces labeling noise, which limits the reliability of MOS prediction. Preference prediction reduces this variability as listeners compare signals directly, producing cleaner labels. We study MOS-free preference prediction and propose PrefSQA, which incorporates uncertainty-aware logits, an impairment attention head, and a module based on non-matching-reference comparisons. We use and refine five datasets, including MOS-derived and low-noise simulated sets with matching and non-matching content, experiment with human preference sets, and test on unseen data. Experiments show small improvements on MOS-derived data, while other sets reveal clear improvement over the baselines, highlighting the value of high-quality preference data and demonstrating the effectiveness of the proposed method.
arXiv:2604.00626v4 Announce Type: replace Abstract: As Large Language Models continue to grow in both capability and cost, transferring frontier capabilities into smaller, deployable students has become an important engineering problem, and knowledge distillation remains a common technique for this transfer. The prevailing recipe in industrial pipelines, static imitation of teacher-generated text, carries a structural weakness that grows more severe as tasks become longer and more reasoning-intensive. Because the student is trained on flawless teacher prefixes but generates its own at inference, small errors tend to accumulate into trajectories it has rarely been trained to recover from, and the resulting exposure bias has been shown to scale roughly with the square of sequence length. On-Policy Distillation reorganizes the training loop around this observation by having the teacher provide feedback on what the student actually produces, with the goal of reducing the compounding term toward linear and reframing distillation as an iterative correction process rather than single-pass imitation. The resulting literature has expanded along divergence design, reward-guided optimization, and self-play, yet contributions remain scattered across the knowledge distillation, RLHF, and imitation learning communities without a unified treatment. This survey provides such a treatment. We formalize OPD as f-divergence minimization over student-sampled trajectories, organize the field along three design axes (what to optimize, where the signal comes from, and how to stabilize training in practice), and consolidate success conditions, recurring failure modes, and the connection between OPD and KL-constrained reinforcement learning. We close with open problems that emerge from this synthesis, including distillation scaling laws, uncertainty-aware feedback, agent-level distillation, and the growing overlap between knowledge distillation and RL.