Numerical Analysis
Integral chemical reaction neural networks
Abraham Reyes-Velázquez and Stefan Güttel
Discovering the structure and kinetics of chemical reaction networks (CRNs) from time-series concentration data is a fundamental challenge in chemical kinetics, with existing approaches relying on prior mechanistic assumptions or suffering from high computational cost and noise sensitivity. In this work, we present integral chemical reaction neural networks (iCRNNs), a framework that combines the interpretable, physics-constrained, architecture of chemical reaction neural networks (CRNNs) with an integral collocation formulation of the governing dynamics. Rather than solving ODEs forward in time at each training step, we approximate the integral of the learned rate functions directly using numerical quadrature. This yields an entirely algebraic forward pass consisting only of
matrix operations, eliminating the repeated adaptive ODE solves of standard CRNN and producing smoother, more predictable training. We provide a recovery error analysis characterising the structural sources of ill-conditioning (conservation laws, reaction reversibility, and shared reactant sets) that limit network identifiability. On two
benchmark CRNs, iCRNN trains in roughly half the wall-clock time of the baseline CRNN method on a four-species mechanism and between three and four times faster on a five-species mechanism, while completing every training run reliably and attaining comparable or lower loss.
Read the full abstract ▾
Cryptography and Security
TERMon: Detecting Persistent Behavioral Threats in Edge AI via Hardware-Native Ternary Runtime Monitor
Arish Sateesan and Edlira Dushku
Edge AI accelerators are increasingly deployed in safety-critical environments, where model outputs may control physical actuators, make access-control decisions, or trigger alarms. In these settings, runtime failures often remain undetected because model corruption, distribution shift, and
adversarial inputs can still produce well-formed, confident predictions. This paper presents TERMon, a lightweight hardware runtime monitor that detects such anomalies by observing inference behavior rather than re-executing or formally verifying the model. TERMon represents class-conditional trusted behavior as hardware-efficient ternary patterns that are matched in parallel against a thermometer-encoded fingerprint. The ternary encoding reproduces the corresponding unquantized range decision exactly. TERMon detects harmful weight corruptions in proportion to their behavioral impact, while out-of-distribution and
adversarial inputs are largely not separable using the monitored features at a strict false-positive operating point. We implemented TERMon on a PYNQ-Z2 FPGA, and the pipelined design requires no on-chip block RAM or DSPs and has a two-cycle decision latency.
Read the full abstract ▾
Computer Vision
ZYT-World: A Real-Time Controllable World Model for Closed-Loop Autonomous-Driving Simulation
Boni Hu, Xiong Wei, Haoming Huang et al. (+20 more)
Generative world models offer controllable and repeatable closed-loop simulation for end-to-end and
vision-
language-action driving policies, but production deployment exposes three unresolved requirements: faithfully reproducing a mixed fisheye-pinhole rig at native resolutions; reconciling causal, per-timestep interaction with long-horizon stability and low latency; and preserving scene identity when a location is revisited. We present ZYT-World, a single architecture that natively generates four fisheye views with field of view > 180{\deg} and three pinhole views. Projection-specific Plucker adapters encode camera geometry, ego-motion adaptive layer normalization provides global motion control, and a lightweight pixel-aligned layout conditions traffic participants and signals through instance-level boxes, headings and colors. Heterogeneous training combines full-rig
geometric coverage with high-resolution detail. Teacher forcing, causal consistency
distillation, self-rollout distribution matching
distillation, and RigCritic transform a 40-step bidirectional teacher into a one-step, per-latent streaming generator, with RigCritic evaluating the seven-view rig jointly. A 19M-parameter variational autoencoder decoder (TinyVAE), W8A8 quantization, and our inference engine reduce decoding, backbone, and incremental-execution costs, respectively. Finally, cross-trajectory pairs derived from real captures train a plug-in implicit-
memory module that preserves place-specific evidence. On the internal multi-view test set, the one-step model retains more than 90% of the teacher's PSNR and SSIM, while FID, FVD, and LPIPS stay within 11% of the teacher. Under the generator-only timing in Figure 2, it is 107.7 times faster than the 40-step bidirectional teacher. TinyVAE decodes 59.8 times faster than Wan. 30s rollouts and cross-trajectory revisits show the intended long-horizon and
memory behavior.
Read the full abstract ▾
All authors Boni Hu, Xiong Wei, Haoming Huang, Yong Huang, Chenbo Wang, Yi Yang, Jiancheng Wang, Ruicheng Zhu, Zhimin Yang, Guanglai Liu, Qiaowan Jin, Dongzhuo Wang, Haiwei Kuang, Jiajun Fan, Yue Wu, Jiaxin Wei, Hao Sun, Feihong Yan, Yuyao Zhou, Wei Bi, Kaixuan Wang, Zichao Guo, Xiaozhi Chen
Computer Vision
SignGPT: Toward LLM-Mediated Sign Language Interaction through Gloss-Free Translation and Generation
Ronghui Li, Jun Dong, Zhongyuan Hu et al. (+8 more)
Large
language models (LLMs) provide limited support for sign
language interaction. Unifying sign
language translation (SLT) and generation (SLG) to enable sign
language as both input and output can reduce switching between separate models during sign-text interaction. We present SignGPT, a unified, pose-based framework for gloss-free SLT and SLG. SignGPT integrates part-aware hierarchical representations of body, hand, and facial motion into a shared
language model and employs asymmetric multi-
token prediction and progressive training for bidirectional modeling. We evaluate SignGPT on How2Sign (ASL) and Phoenix-2014T (DGS) through
benchmark comparisons, qualitative analyses, and component ablations. An exploratory study with 12 Deaf ASL signers assesses an LLM-mediated sign-to-sign response pipeline, highlighting the potential of unified modeling to support sign
language conversation (SLC).
Read the full abstract ▾
All authors Ronghui Li, Jun Dong, Zhongyuan Hu, Zunnan Xu, Jun Zhou, Liyuan Chen, Shuoling Liu, Jiangpeng Yan, Jie Guo, Xiu Li, Linchao Bao
Robotics
NeuRIO: A Streaming Neural Estimator for Zero-Shot Sim-to-Real Multi-Robot Relative Inertial Odometry
Zhehan Li, Jiadong Lu, Shengwei Ren et al. (+2 more)
We present NeuRIO, a streaming neural
estimator for anchor-free 6-DoF relative inertial odometry using only identified inter-
robot bearings, ranges, and IMU measurements. NeuRIO canonicalizes measurements into
gravity-aligned coordinates, represents
robots as nodes and mutual observations as factors, and uses
attention for spatial
reasoning and GRUs for temporal modeling. As a graph network, NeuRIO applies shared node-wise and factor-wise operators throughout the network, enabling it to handle different team sizes and time-varying observation graphs. NeuRIO is trained on a simulator that couples various motion patterns, device-level sensor characteristics, and diverse, realistic modeled, and temporally persistent sensor corruptions. In this way, NeuRIO achieves zero-shot sim-to-real transfer. Across $24$ real-world sequences, NeuRIO achieves $14.1\,\mathrm{cm}$ position RMSE and $3.9^\circ$ rotation RMSE. More importantly, NeuRIO demonstrates strong computational scalability, maintaining an update cost below $20\,\mathrm{ms}$ with up to $400$
robots in simulation, while optimization-based methods exceed $20\,\mathrm{ms}$ at only $24$
robots. Moreover, even trained on limited team sizes, NeuRIO transfers directly to unseen larger teams without architectural or parameter changes.
Read the full abstract ▾
All authors Zhehan Li, Jiadong Lu, Shengwei Ren, Chao Xu, Yanjun Cao
Machine Learning
SpecQuant: Speculative Decoding with Multi-Parent Quantization for Adaptive LLM Inference
Harish KB, Jagadeeswaran M, Pradheep P et al. (+2 more)
Running large
language models (LLMs) locally continues to be limited by restrictions of compute and
memory on consumer hardware. The popular acceleration technologies, such as quantization, speculative decoding, and adaptive inferencing, offer substantial speed boosts but usually necessitate retraining, per architecture tuning, or draft models. SpecQuant is a trainingfree framework, that combines speculative decoding with multiparent quantization to perform adaptive, efficient inference of LLMs. SpecQuant derives multiple quantized variants (INT4, FP8, FP16) from a shared base model, and dynamically routes queries based on predicted complexity; lightweight variants are used for simple or factual tasks, and full-precision models are used for complex
reasoning tasks or long-context inputs. The shared-weight design of SpecQuant ensures sufficient
token acceptance for speculative decoding without compatibility issues using separate draft parent models. We evaluate SpecQuant on Qwen2.5 based models on the MMLU, AlpacaEval, and GSM8K
datasets, or
benchmarks, demonstrating 35-43% speedups without degrading accuracy greater than 2%, substantial within the LLM community. SpecQuant enables practical on-device LLM deployment across diverse hardware without special infrastructure or expertise.
Read the full abstract ▾
All authors Harish KB, Jagadeeswaran M, Pradheep P, Yuvanesh S, Sivakumar T
Discrete Mathematics · Cross-listed from q-bio.PE
Best Matches in Phylogenetic Networks
Patricia A. Ebert, Peter F. Stadler and Marc Hellmuth
Best match graphs (BMGs) were introduced in mathematical phylogenetics to describe the concept of closest relatives for related genes (leaves of rooted tree) in different organisms (defining leaf colors). We generalize this concept here to leaf-colored rooted networks, where least common ancestors are in general neither unique nor comparable. We characterize BMGs of rooted networks as those vertex-colored digraphs that are properly colored and satisfy an easy-to-check condition that we call the sicor-in-hub property. BMGs can be recognized in linear time and an explaining network can be constructed in quadratic time. Analogous results are obtained for reciprocal best match graphs (RBMGs), where an edge $\{x,y\}$ corresponds to pairs of vertices with different color that are mutually closest relatives.
Read the full abstract ▾
Computer Vision
Diffusion-Based Tumor Inpainting for Renal Segmentation under Clinical Data Scarcity
Ekaterina Sedykh, Salme Ussanov, Dmytro Fedorenko et al. (+1 more)
Deep learning
segmentation of renal tumors requires large annotated
datasets, yet clinical deployments typically offer only a handful of tumor-positive cases from the target site. We propose a
diffusion-based inpainting framework that synthesizes anatomically plausible renal tumors within healthy CT scans, requiring no additional annotation, and provide the first systematic comparison of 2D, 2.5D, and full 3D (MAISI) synthesis strategies for this task. Training the
diffusion model on public data (KiTS23, KIRC) and evaluating nnU-Net
segmentation on a internal cohort across three low-data regimes, we find that 2.5D and 3D augmentation substantially reduce false positives (from $\sim$18–20% to $\sim$3–6%) while maintaining Dice, whereas 2D provides no consistent benefit. Crucially, the proposed 2.5D method matches full 3D synthesis on every metric at substantially lower computational cost, indicating that local volumetric consistency alone is sufficient for effective augmentation in data- and resource-scarce clinical settings.
Read the full abstract ▾
All authors Ekaterina Sedykh, Salme Ussanov, Dmytro Fedorenko, Dmytro Fishman
Hardware Architecture
Integrating Approximate Logic Synthesis into Approximate High-Level Synthesis
Jian Shi, Ruicheng Dai, Chang Meng et al. (+2 more)
Approximate high-level synthesis (HLS) and approximate logic synthesis (ALS) are two techniques for generating approximate
circuits. They operate at different granularities. Approximate HLS typically modifies instructions in a control and data flow graph, whereas ALS modifies gates and interconnects in a gate-level netlist. The absence of a unified framework combining these techniques limits the potential for joint optimization. To bridge this gap, we propose to integrate ALS into the flow of approximate HLS. This integration expands the design space of approximate HLS by introducing fine-grained approximation induced by ALS, thereby generating approximate
circuits with higher quality. Experimental results show that under the same error bound, our method reduces the hardware cost by 11% on average compared to the state-of-the-art methods.
Read the full abstract ▾
All authors Jian Shi, Ruicheng Dai, Chang Meng, Yue Yang, Weikang Qian
Data Structures and Algorithms
Submodular Maximization over Bipartite Perfect Matchings and Matroid Intersection Bases
Chandra Chekuri, Lars Rohwedder, Neta Singer et al. (+2 more)
Motivated by applications in
fairness and foundational questions, we consider the problem of maximizing a monotone submodular function $f\colon 2^E \rightarrow \mathbb{R}_+$ over maximum cardinality sets in the intersection of two matroids on a common ground set $E$. An important special case is submodular perfect matching in bipartite graphs. Prior to this work, its approximability was poorly understood with only constant inapproximability known, despite not even a $\frac{1}{o(\sqrt{|E|})}$-approximation being known. Even when allowing to violate the cardinality constraint slightly, only a bicriteria approximation with a significant loss in the objective was known. Here, we obtain two results. First, we show that, within constant factors, the problem is approximation-equivalent to Submodular Orienteering in directed graphs. This yields an $\Omega(1 / \log |E|)$-approximation in quasi-
polynomial time together with an almost-matching hardness result. Second, we obtain an improved
polynomial-time bicriteria approximation via a local search framework. More precisely, if $f(T^*)$ is the largest submodular value of a common independent set in both matroids of size at least $K$, we find a common independent set $T$ such that $|T| \geq (1 - \epsilon) K$ and $f(T) \geq (1/2 - \epsilon) f(T^*)$. In contrast, previous work only guarantees a value of $\Omega(\epsilon) f(T^*)$ while ensuring that $|T| \geq (1 - \epsilon) K$.
Read the full abstract ▾
All authors Chandra Chekuri, Lars Rohwedder, Neta Singer, Jan Vondrák, Rico Zenklusen
Machine Learning · Cross-listed from astro-ph.IM
Bayesian classification of astronomical spectra with class uncertainties
Simon Barton, Martin Sahlén, Andreas Korn et al. (+1 more)
Context: We developed a probabilistic machine learning method with the aim of performing the O(10)-way classification of low- and high-resolution spectra of
stellar and extragalactic targets for the upcoming 4MOST survey. In fulfilment of the survey requirements, this method should be able to express uncertainty in the input data as well as uncertainty introduced in its prediction. Aims: Four different methods are explored: (1) convolutional neural networks (CNNs), (2) the Dirichlet distribution, (3) Monte Carlo dropout (MCD), (4)
Bayesian neural Networks (BNNs) + variational inference (VI). Training and validation was performed using labelled spectra from the SDSS database and a custom 4MOST mock
dataset. All the methods were compared in terms of the same metrics: accuracy, area under the curve (AUC), expected calibration error (ECE), Shannon
entropy, negative log-likelihood (NLL), Brier score, training time, and inference time. Methods: A CNN with simple architecture and about 20,000 parameters was trained to achieve classification accuracies of 91.5% on SDSS data and 92.8% on 4MOST mock data. The direct Dirichlet prediction and VI models tested provide uncertainties on class membership probabilities, but they confuse classes more often. The MCD on a CNN is found to be the most suitable; it boosts the point-estimate accuracies to 92.6% and 93.9%, while still providing fast training and sufficiently fast inference. Compared to a standard CNN, the method additionally provides well-calibrated uncertainties at marginal extra cost.
Read the full abstract ▾
All authors Simon Barton, Martin Sahlén, Andreas Korn, Christian Glaser
Machine Learning
Optimization Geometry of Equivalent Brownian RKHS Representations
Mahdi Mohammadigohari and Gustau Camps-Valls
Equivalent finite parameterizations can represent the same functions and intrinsic norm yet induce different optimization
algorithms. We study this effect in a controlled finite Brownian RKHS with nodal, increment, and spectral coordinates. Classical finite-element, RKHS-interpolation, Brownian-covariance, and mixed-boundary DCT identities make the shared hypothesis class, Brownian energy, approximation operator, and coordinate maps explicit. Our main results concern the optimization geometry of this fixed model. With mapped initialization, identical scalar steps, and identical minibatches, nodal and spectral GD/SGD have exactly the same mapped trajectories. Increment GD is an explicit Euler step for the constant Brownian/Sobolev metric, with factor $1/h$. For Brownian-regularized least squares, $\kappa_2(\mathbf H_{\mathrm{inc}})\le1+A/\rho$, independently of grid resolution $G$ for fixed $A$, $\rho>0$, and the stated normalization. Under the stated standard-Adam convention, the universal orthogonal equivariance group is exactly the signed permutations; the block DCT-VIII transform is not one. Float64 tests over five grids numerically verify the finite identities, mapped one-layer and recursive trajectories, conditioning predictions, and
theorem-matched Adam separation. Thus coordinate effects are isolated without changing the represented functions, intrinsic regularizer, or approximation space.
Read the full abstract ▾
Robotics
RAYA: Learning Where and When to Intervene for Robot Recovery
Ishaan Mahajan, Charles Chen, Frederike Dümbgen et al. (+1 more)
A
robot can predict failure and still be unable to prevent it. By the time a safety mechanism reacts, the nominal plan may already have spent the control authority that recovery requires, and fixed task priorities may block whatever response remains. Our key insight is that both aspects are decided inside the controller. Recoverability must inform actions while they are chosen rather than veto them afterward, and task objectives must be adapted as recoverability shrinks. Building on this, we present RAYA, a hybrid learned-analytic framework that places a learned finite-horizon recoverability margin inside an optimal controller with hard constraints and pairs it with a bounded learned scheduler that shifts task weights to facilitate recovery. Across 7,200 simulation episodes per controller spanning quadrotor and autonomous-vehicle
benchmarks, RAYA not only improves survival rates, but also transfers the learned components zero-shot to unseen trajectories, disturbances, plant shifts, and friction layouts. We developed an embedded realization of RAYA and deployed it on-board a 35g Crazyflie quadrotor. Across 40 combined hardware flights under wind with either aerodynamic mismatch or an unmodeled 40% motor-command loss, each of three baselines fails in all trials, while RAYA completes 10/10 six-cycle missions. Project Website: https://raya-control.github.io/.
Read the full abstract ▾
All authors Ishaan Mahajan, Charles Chen, Frederike Dümbgen, Brian Plancher
Cryptography and Security
CIPL: A Channel-Aware Framework for Recoverable Privacy Leakage in LLM Agents
Tao Huang, Guosen Wu, Guolong Zheng et al. (+5 more)
Privacy leakage in LLM
agents is commonly evaluated within individual components such as
memory,
retrieval, or tool-use pipelines, which makes it difficult to distinguish internal exposure from information that an external observer can actually recover. We present CIPL (Channel Inversion for
Privacy Leakage), a channel-aware evaluation framework for black-box
privacy leakage in LLM
agents. CIPL represents a target through sensitive source, selection, assembly, execution, observation, and extraction stages and evaluates the transition from selected sensitive units to attacker-recoverable output under a shared protocol. Experiments across
memory-based,
retrieval-mediated, and tool-mediated targets, together with a BrowserUse live-
agent case study, show that storage labels alone do not determine recoverability.
Memory targets form a near-saturated reference case,
retrieval-mediated leakage is frequently partial, and tool-mediated and live-
agent leakage varies strongly with observation surface,
prompt-to-channel
alignment,
retrieval depth, and provider behavior. A stratified semantic audit further identifies attacker-useful disclosures that canonical exact matching misses. CIPL therefore provides a common framework for comparing how internal sensitive dependence is realized as externally recoverable leakage across heterogeneous
agent pipelines.
Read the full abstract ▾
All authors Tao Huang, Guosen Wu, Guolong Zheng, Jiayang Meng, Chen Hou, Xu Yang, Xuechao Yang, Feng Xia
Artificial Intelligence
Listen Before You Speak: Response Planning from Listener Facial Reactions for Conversational Speech Generation
Yunji Chu
Conversational
speech depends on dialogue context and the listener's immediately preceding behavior. We propose ReACT-TTS, a two-stage framework that uses a one-second pre-response listener facial sequence to plan the next utterance's
emotion and prosody before
speech realization. On a strict dyadic MELD protocol, Temporal conditioning yields higher mean macro-F1 and VAD concordance than Text-only across ten seeds, while accuracy remains essentially unchanged. Ablations show that temporal modeling performs best among the visual variants and that an explicit early-to-late difference is unnecessary; correct listener reactions also outperform cyclic mismatches on average. In a contextual-appropriateness study with 20
speech researchers, 76% of judgments prefer Temporal, 9% Text-only, and 15% report no preference. We further connect the predicted response style to a Grad-TTS backbone for end-to-end
speech realization. Overall, the results support pre-response listener dynamics as complementary cues for conversational response planning. The source code is available at https://github.com/CYJ1/ReACT-TTS_public.
Read the full abstract ▾
Performance
Performance Analysis of Low-Order, GPU-accelerated Finite Element Kernels using Kokkos
Fabian Böhm, Nils Kohl, Harald Köstler et al. (+1 more)
We study performance portability for low-order,
matrix-free finite element kernels, using the example of a vectorial, variable-coefficient PDE operator originating in geophysical models. Written in Kokkos, the kernel is compared on NVIDIA H100, AMD MI250X, AMD MI300A and Intel PVC Max 1550 GPUs. Owing to its low order and to optimizations that reduce the arithmetic, the kernel has a low arithmetic intensity, so that its performance is determined by how the finite element assembly is mapped onto the
memory hierarchy. This is a dimension in which the architectures differ even within one vendor family, causing different performance characteristics. We examine how Kokkos' hierarchical parallelism and shared scratch
memory, which are used for the shared degrees of freedom of the conforming discretization, behave on each device. Finally, we show how portability gaps can be narrowed with tuning levers such as the size of the thread groups, the balance between occupancy and register use, and the atomic accumulation strategy at the end of the kernel.
Read the full abstract ▾
All authors Fabian Böhm, Nils Kohl, Harald Köstler, Ulrich Rüde
Artificial Intelligence
GUARD: Natural Forgetting in Large Reasoning Models via Guided Answer-Reasoning Distillation
Zeyu Yan, Guanghao Zhou, Minghui Qiu et al. (+2 more)
Recent advances in large
reasoning models (LRMs) have made machine unlearning more challenging, as protected facts or unsafe rationales may surface in intermediate chain-of-thought (CoT) traces before the final answer is produced. Existing unlearning objectives typically suppress the target content or redirect internal representations, but they never specify how the post-forgetting trajectory should continue, which can lead to hallucinated substitutes, malformed boundaries, or repetitive outputs. We argue that LRM unlearning should instead learn a natural forgetting trajectory: a coherent non-disclosing CoT followed by a stable refusal-style answer that replace the original disclosure. To this end, we propose Guided Answer-
Reasoning Distillation (GUARD), which converts model-generated unsafe disclosures into safe-exit trajectories, aligns a frozen LRM via guidance
tokens, and distills the guided behavior into model parameters. To address the lack of metrics for replacement quality beyond leakage, we further introduce Natural Forgetting
Reasoning Score (NFRS), which captures structural stability, fluency, and unsupported substitutes in forgotten outputs. Extensive experiments on R-TOFU and a
STAR-1-derived harmful-intent setting show that GUARD substantially reduces unsafe and
privacy disclosures across two widely adopted distilled LRMs while preserving
reasoning utility.
Read the full abstract ▾
All authors Zeyu Yan, Guanghao Zhou, Minghui Qiu, Ming Gao, Cen Chen
Computation and Language · Cross-listed from eess.AS
The Spoken Wikipedia Presentation Corpus
Thomas Ranzenberger, Steffen Freisinger, Tobias Bocklet et al. (+1 more)
We present the Spoken Wikipedia Presentation Corpus, an extension of the Spoken Wikipedia Corpora featuring LLM-generated slide decks for
multimodal ASR. Slides are created from LLM-segmented sections using a hybrid pipeline that combines LLM-based content planning with rule-based design decisions. For each section, an LLM generates a slide title, bullet points, a takeaway message, and a visual description that is used to create an illustration. Rule-based matching then selects layouts, themes, and styles to produce the final slides. A
vision LLM extracts slide text as Markdown. We evaluate multiple ASR and spoken
language models (SLMs). The best model achieves an average micro-WER of 10.23% and an average micro-CER of 6.48% on audio-only inputs. English yields the lowest error rates, followed by German and Dutch, while performance declines across lower-resource
languages. Although audio-only baselines are strong,
multimodal zero-shot prompting of omni models remains challenging. The aligned slide, text, and audio data show a strong potential to improve recognition through cross-modal context.
Read the full abstract ▾
All authors Thomas Ranzenberger, Steffen Freisinger, Tobias Bocklet, Korbinian Riedhammer
Computer Vision
Wan Xu, Yuanfan Guo, Kevin Han et al. (+2 more)
Despite the remarkable progress in
Multimodal Large
Language Models (MLLMs), prevailing Chain-of-Thought (CoT) paradigms remain confined to the natural-
language expression space. Consequently, they inherently incur excessive linguistic overhead, leading to information dilution and weak visual grounding. To address this challenge, we propose Dense
Reasoning Trace (DRT), a paradigm that departs from natural-
language-centered CoT by expressing
reasoning as compact structured traces, which include concise intermediate states with symbolic connectors and disentangle visual observations from logical deductions. First, we introduce the Dense Trace Initialization to internalize the DRT
reasoning mode into the model, substantially improving
token efficiency while preserving visual evidence. To further enable the model to faithfully capture the logical relations within traces, we propose the Trace-Grounded Reinforcement Learning framework, which builds reference traces through a tri-perspective verification pipeline and employs Trace-Grounded GRPO with structured rewards, encouraging the model to generate concise DRT-style traces with reduced
hallucination and stronger logical grounding. Extensive experiments on challenging
reasoning benchmarks show that DRT achieves 5.5$\times$
token efficiency improvement while improving 1.3 accuracy points over the Qwen3-VL baseline. These findings suggest that complex
multimodal reasoning may not require verbose natural-
language traces, opening a more efficient path for next-generation MLLMs. Our code and data are available at: https://github.com/HIT-leaderone/DRT
Read the full abstract ▾
All authors Wan Xu, Yuanfan Guo, Kevin Han, LaLa Chen, Wangmeng Zuo
Computation and Language
PRISM-BN: A Controlled Corpus and Benchmark for Text-to-Parameterized Bayesian Network Extraction
Amartya Bhattacharya, Nikhil Singh, Neeti Pokhriyal et al. (+1 more)
Probabilistic Graphical Models (PGMs), especially
Bayesian Networks (BNs), expose directed structure and probabilistic parameters, making them natural symbolic targets for neurosymbolic AI. Yet training text-to-parameterized-BN systems requires paired text-to-BN resources unavailable at scale. We introduce PRISM-BN, a controlled corpus of 5054 BN-grounded descriptions paired with discrete reference BNs containing variables, states, directed edges, root priors, and full multi-parent CPDs across five domains. The instances are derived from 50 Wikipedia-seeded backbones, and their probabilities are internally constructed
benchmark targets rather than externally validated causal estimates. PRISM-BN is built with PRISM, a marginal-first pipeline that elicits marginal and local joint distributions, analytically recovers normalized CPDs, and constructs locally reparameterized subgraphs. We define a
benchmark with semantic node and state
alignment, conditional structural scoring, and strict full-CPD evaluation. Across six LLM extractors, Node F1 ranges from 0.56 to 0.83, conditional Edge F1 from 0.90 to 0.97, and CPD-KL from 1.11 to 3.14. Conditional state and edge recovery remain consistently strong, whereas strict full-CPD agreement remains challenging. These trends persist with independently generated GPT-5.5 references, and a human pilot corroborates structural recoverability and similar probabilistic interpretations. PRISM-BN supports separate evaluation of structural recovery and probabilistic parameter estimation.
Read the full abstract ▾
All authors Amartya Bhattacharya, Nikhil Singh, Neeti Pokhriyal, Soroush Vosoughi
Artificial Intelligence
Accelerating Dense LLMs via L0-regularized Mixture-of-Experts
Zhenyu Zhang, Jiudong Yang, Zhaowen Tao et al. (+1 more)
Large
language models (LLMs) achieve strong performance but suffer from slow and costly inference. Existing acceleration methods often lead to noticeable performance degradation, while Mixture-of-Experts (MoE) models require extensive computational resources. In this paper, we propose L0-MoE, a lightweight MoE approach using L0-regularization to accelerate dense LLMs nearly without performance loss. Our method introduces a cluster confusion
matrix for domain-aware
dataset curation and applies dynamic batching for efficient training. Experiments show that L0-MoE achieves up to 2.5x speedup over dense models while maintaining competitive performance, outperforming existing LLM acceleration baselines.
Read the full abstract ▾
All authors Zhenyu Zhang, Jiudong Yang, Zhaowen Tao, Meng Chen
Digital Libraries
From Code Archival to Knowledge Graph: Bridging Software Heritage, COAR Notify and Wikidata
Camillo Carlo Pellizzari di San Girolamo and Francesco Tosoni
Software is a first-class scientific object, yet validated links between source code and the scholarly record remain largely absent from the Linked Open Data (LOD) cloud, isolating archived artefacts from semantic discovery. This paper presents an end-to-end reconciliation pipeline that harvests, validates, and models publication-to-repository pairs from sources where the link between a paper and its source code is explicit and editorially verified: the software-centric journals JOSS, SoftwareX, and IPOL, together with the reproducibility reports of the SIGMOD Availability and Reproducibility Initiative (ARI). This yields a curated corpus of 4,397 $\langle$DOI, repository-URL$\rangle$ pairs. We design two distinct application profiles grounded in Wikidata classes (one for scholarly articles, one for software instances) aligned with the schema.org and CodeMeta vocabularies. This architectural separation enables rule-based reconciliation at two granularities: lightweight, inline publication references or standalone, first-class Wikidata software nodes equipped with SWHIDs, Software Heritage's content-addressed identifiers. A read-only lookup against Wikidata shows that only 82 of the harvested repositories were already modelled there; human-reviewed batches have since created 4{,}182 new software items cross-linked to their articles. We further show that payloads of the emerging COAR Notify protocol, an external effort we do not develop, map natively onto our input format, so the same backend could later serve a live enrichment stream. Our core contribution is a pair of application profiles that turn Wikidata into a connector between the scholarly record and archived source code; we openly release all code, application profiles, and harvested
datasets.
Read the full abstract ▾
Artificial Intelligence · Cross-listed from eess.AS
Samsone: A Family of Open Small Audio Language Models for On-Device Inference
Piotr Masztalski, Michał K. Grzeszczyk and Olaf Sikorski
The success of Large Audio
Language Models has driven the development of massive
multimodal networks exceeding billions of parameters. However, the demand for
privacy-preserving, low-latency processing has shifted focus toward Small Audio
Language Models (SALMs) capable of on-device execution. In this paper, we introduce Samsone, a family of SALMs designed for edge computing. Our core model, Samsone-134M, establishes a new state-of-the-art for its size class across multiple
benchmarks. We further explore the scaling laws of SALMs by introducing Samsone-99M and Samsone-356M. Despite their compact footprint, the Samsone family delivers performance competitive with models orders of magnitude larger. To foster open research and reproducibility, we train Samsone on publicly available data. We release the training code, model weights, mobile-optimized checkpoints and provide an open-source Android application to demonstrate real-time on-device inference of Samsone.
Read the full abstract ▾
Machine Learning
Multi-Domain Clustering via Measure Quantization
Rafael Pereira Eufrazio, Eduardo Fernandes Montesuma and Charles Casimiro Cavalcante
Clustering is a fundamental task in data analysis, typically addressed through centroid-based methods such as K-means. In this work, we present a general framework for multi-domain clustering via measure quantization: given samples from multiple domains, we learn a shared set of cluster prototypes by minimizing a probability metric, such as the Sinkhorn divergence or the Maximum Mean Discrepancy, between each domain's probability measure and the measure of prototypes. Data points are then assigned to clusters either via nearest centroid, or via optimal transport, a collaborative strategy that couples all samples within a domain. A mini-batch optimization strategy makes both fitting and assignment scalable, reducing
memory and computational cost while preserving clustering performance. Experimental results on 5 multi-domain
benchmarks spanning image, audio and sensor data show that our Sinkhorn-based method consistently outperforms classical and multi-domain clustering baselines, and that this advantage persists when scaling to hundreds of thousands of samples.
Read the full abstract ▾
Computation and Language
Rethinking Human-Aligned Evaluation: An Analysis of Semantic Metrics Beyond WER
Hritika Sharma, Thibault Bañeras-Roux, Alessandra Pinto et al. (+4 more)
Word Error Rate (WER), the most commonly used metric for Automatic
Speech Recognition (ASR), treats every lexical deviation from the reference as equally costly, regardless of whether it changes meaning. This raises the question: does WER actually track how humans judge ASR transcript quality? We introduce HATS-en, an English
dataset for human-centered ASR evaluation. Using this
dataset, we
benchmark lexical metrics against several configurations of BERTScore and SemDist, varying the
language model, layer, and pooling strategy. We find that WER agrees least with human judgment among all metrics tested, that the best-performing SemDist configurations achieve the highest overall agreement, ahead of CER and BERTScore, and that no single model is best across settings. CER, despite its simplicity and low cost, remains remarkably close to these best configurations. In line with prior recommendations, our results support shifting ASR evaluation toward CER both for English and for morphosyllabic writing systems as it is a more interpretable and low-cost metric for what evaluation should actually capture, and using SemDist as a complementary evaluation.
Read the full abstract ▾
All authors Hritika Sharma, Thibault Bañeras-Roux, Alessandra Pinto, Petr Motlicek, Hyunggu Jung, Esaú Villatoro-Tello, Somang Nam
Computation and Language
When Steering Fails in Latent Reasoning: A Latent-to-Language Transition Gap
Gaoxiang Huang and Lei Qi
Activation steering has become a widely used approach for controlling
language models during explicit chain-of-thought (CoT)
reasoning, motivating its extension to latent CoT. However, we find that steering continuous thoughts produces substantially weaker effects on subsequent
language generation than steering explicit CoT, even when the hidden representations are moved by comparable amounts. We first show that task information remains identifiable in continuous thoughts. Hence, we hypothesize a latent-to-
language transition gap, in which an intervention effect in latent space fails to transfer to
language generation. Two further results support this hypothesis: the output distribution changes abruptly at the transition boundary, and task-related directions exert much weaker bidirectional control in latent CoT than in explicit CoT. These findings identify the transition interface as a central target for evaluating and designing future latent-steering methods.
Read the full abstract ▾
Robotics
Outcome-Conditioned End-Effector Geometry Across Vision-Language-Action Policies
Xingyu Lin, Zhuang Li, Zhongrun Wu et al. (+2 more)
Vision-
language-action (VLA) policies solve the same manipulation task through different action interfaces, but task success alone does not establish whether their physical executions agree. We study cross-policy end-effector geometry in 15,000 closed-loop LIBERO rollouts from four policies. The primary clean-condition analysis forms 3,600 configuration-matched, and therefore dependent, policy pairs. Both-success pairs have a median normalized dynamic time warping distance of 0.0120 m versus 0.0380 m when exactly one policy succeeds. This ordering holds in every task, every policy pair, and nine sampling and band-limited representations; however, the ratio varies severalfold across representations, so we report the direction rather than a fixed multiple. Both-failure pairs are more separated again but rest on thin, uneven support, so we report them as exploratory. Within successful executions, partner replacements separate more across tasks than across initial states. A matched baseline still reveals measurable, heterogeneous residual policy differences, so a low cross-policy distance does not imply interchangeability. Successful executions sit about as far from same-task demonstrations as those demonstrations sit from each other, compatible with task-associated geometry without separating training-data overlap from task constraints. A common 72-action window preserves the ordering but reduces its magnitude; endpoint and duration adjustment likewise leaves a positive mixed-outcome coefficient relative to both-success pairs, though its magnitude is specification-dependent. Under composite visual stress, policy rankings and pair composition change together.
Read the full abstract ▾
All authors Xingyu Lin, Zhuang Li, Zhongrun Wu, Shouquan Zhou, Dehui Du
Machine Learning
Beyond Gaussian Worlds: Latent Geometry Matters for JEPAs
Léo Nicollier, Enric Meinhardt-Llopis, Marc Pic et al. (+2 more)
Recent Joint-
Embedding Predictive Architectures (JEPAs) prevent representation collapse by constraining learned representations to follow a prescribed target distribution, such as an isotropic Gaussian or the uniform distribution on a hypersphere. Klindt et al. (2026) showed that, under their Euclidean assumptions, matching a Gaussian target can recover Gaussian latent variables up to a linear transformation, and that the Gaussian is the unique distribution with this guarantee. We extend their analysis to latent variables supported on embedded Riemannian
manifolds and derive conditions on the latent geometry and positive-pair dynamics under which
alignment and exact distribution matching guarantee linear recovery. In particular, when the latent variables are uniformly distributed on a sphere and the representations are matched to the same spherical distribution, every optimal representation recovers the latent state up to an orthogonal transformation. This shows that Gaussian uniqueness is not a universal property of distribution-matched JEPAs: non-Euclidean latent geometries can admit other linearly recoverable distributions. We further derive an approximate-recovery bound that is strictly tighter for the spherical world than for the Gaussian world. Experiments on Gaussian, spherical, and toroidal latent spaces show that geometrically compatible targets yield better linear recovery when optimization succeeds, whereas mismatched targets distort the latent structure. This advantage persists in high-dimensional Clifford-
torus worlds.
Read the full abstract ▾
All authors Léo Nicollier, Enric Meinhardt-Llopis, Marc Pic, Pablo Musé, Gabriele Facciolo
Computation and Language
Analysing the Linearity of Linguistic Relations in Language Model Embedding Spaces
Vasudevan Nedumpozhimana, Fathima Thekkekara and John Kelleher
We propose a framework to analyse how strongly different linguistic relations are linearly encoded in
language model
embedding spaces. We formalise linear encoding via a constrained linear approximation over related and unrelated word pairs and apply this to an extended BATS
dataset covering inflectional, derivational, lexicographic, and encyclopedic relations in GloVe, RoBERTa, and ModernBERT. Our experiments show near-perfect linear encodings for inflectional and derivational relations, but substantially higher errors for lexicographic and encyclopedic relations, especially for one-to-many and many-to-many associations. We also find that RoBERTa and ModernBERT generally encode relations more linearly than GloVe. These results indicate that our framework can reveal which relational structures are most linearly accessible in
embeddings, offering a compact tool for probing and comparing relational geometry across models.
Read the full abstract ▾
Computer Vision
Configurable Multi-Stage Vision Pipeline for Crop Disease and Pest Diagnosis
Naga Ganesh, Chandrashekar M S, Lakshmi Pedapudi et al. (+2 more)
FarmerChat is Digital Green's farm advisory service for smallholder farmers. When something looks wrong with a crop, the farmer takes a photograph and sends it, and that photograph is the whole question: no symptom described, no crop named, often no text at all. The service has to determine whether the picture can be used, what crop it shows, and what is wrong with it, from images taken on cheap phones in a field, in poor light and with a moving camera. The system doing this today cannot be adjusted. It has no adjustable thresholds for photograph rejection, crops and problems cannot be added, and there is no confidence cut-off to set. We study about 1.16 million photographs sent to FarmerChat from Ethiopia, India, Kenya and Nigeria. The production quality gate rejected 46.8% of the images it judged, over a quarter of those reaching diagnosis returned no crop name, and 35.8% of the labelled problems filed under "disease" are pests, identifiable without the crop. We therefore split the work into three stages: a quality gate (M0), a crop detector (M1), and a disease or pest detector (M2). Route A fills all three with one fine-tuned
vision-
language model (Qwen3-VL-4B) answering in a single call. Route B fills each with a small specialist model (DaViT, YOLO26). We replace our production GPT-4o quality gate with a small MobileNetV3 gate at 86.9% F1 in 12 ms. On one test set scored the same way for every system, a hierarchical DaViT-Base achieves 95.41% crop accuracy against 91.46% for the production baseline. It also leads on diagnosis and never declines to answer, while every
language model in the comparison leaves a large share of rows with no diagnosis. The fine-tuned model retains two capabilities the specialists do not have: one call for all three stages, and a request for a better photograph when the image cannot support an answer.
Read the full abstract ▾
All authors Naga Ganesh, Chandrashekar M S, Lakshmi Pedapudi, Aakash Singh, Vineet Singh