arXiv:2605.13790v1 Announce Type: cross Abstract: Partial differential equations (PDEs) are fundamental for modeling complex natural and physical phenomena. In many real-world applications, however, observational data are extremely sparse, which severely limits the applicability of both classical numerical solvers and existing neural approaches. While neural methods have shown promising results under moderately sparse observations, their inference […]
Deontic Argumentation
arXiv:2509.25781v2 Announce Type: replace Abstract: We address the issue of defining a semantics for deontic argumentation that supports weak permission. Some recent results show that grounded semantics do not support weak permission when there is a conflict between two obligations. We provide a definition of Deontic Argumentation Theory that accounts for weak permission, and we […]
Asymmetric On-Policy Distillation: Bridging Exploitation and Imitation at the Token Level
arXiv:2605.06387v3 Announce Type: replace-cross Abstract: On-policy distillation (OPD) trains a student on its own trajectories with token-level teacher feedback and often outperforms off-policy distillation and standard reinforcement learning. However, we find that its standard advantage weighted policy gradient suffers from three structural weaknesses, including high variance updates, vanishing gradients in zero-advantage regions, and exploration bottlenecks […]
Context Learning for Multi-Agent Discussion
arXiv:2602.02350v3 Announce Type: replace Abstract: Multi-Agent Discussion (MAD) has garnered increasing attention very recently, where multiple LLM instances collaboratively solve problems via structured discussion. However, we find that current MAD methods easily suffer from discussion inconsistency, LLMs fail to reach a coherent solution, due to the misalignment between their individual contexts.In this paper, we introduce […]
The $gamma_c$-Peak: Covariant Recovery on Four Organic Qubit Platforms
arXiv:2605.00026v2 Announce Type: replace Abstract: The Petz recovery map (1986) provably reverses a noisy quantum channel on a reference state, but its algorithmic relevance to real, dissipation-dominated platforms has remained unclear. Using the open-source textttorganic-qc-bench simulation package, we benchmark a Petz-style covariant-purification quantum error correction (CQEC) protocol across four engineered organic qubit platforms operated emphwithout […]
Gyan: An Explainable Neuro-Symbolic Language Model
arXiv:2605.04759v2 Announce Type: replace-cross Abstract: Transformer based pre-trained large language models have become ubiquitous. There is increasing evidence to suggest that even with large scale pre-training, these models do not capture complete compositional context and certainly not, the full human analogous context. Besides, by the very nature of the architecture, these models hallucinate, are difficult […]
EpiGraph: Building Generalists for Evidence-Intensive Epilepsy Reasoning in the Wild
arXiv:2605.09505v2 Announce Type: replace Abstract: Epilepsy diagnosis and treatment require evidence-intensive reasoning across heterogeneous clinical knowledge, including biosignal patterns, genetic mechanisms, pharmacogenomics, treatment strategies, and patient outcomes. In this work, we present textscEpiGraph, a large-scale epilepsy knowledge graph and benchmark for evaluating knowledge-augmented clinical reasoning. textscEpiGraph integrates 48,166 peer-reviewed papers and seven clinical resources into […]
Stylized Text-to-Motion Generation via Hypernetwork-Driven Low-Rank Adaptation
arXiv:2605.13333v1 Announce Type: cross Abstract: Text-driven motion diffusion models are capable of generating realistic human motions, but text alone often struggles to express fine-level nuances of motion, commonly referred to as style. Recent approaches have tackled this challenge by attaching a style injection mechanism to a pretrained text-driven diffusion model. Existing stylization methods, however, either […]
Table-R1: Region-based Reinforcement Learning for Table Understanding
arXiv:2505.12415v3 Announce Type: replace-cross Abstract: Tables present unique challenges for language models due to their structured row-column interactions, necessitating specialized approaches for effective comprehension. While large language models (LLMs) have demonstrated potential in table reasoning through prompting and techniques like chain-of-thought (CoT) and program-of-thought (PoT), optimizing their performance for table question answering remains underexplored. In […]
Ilov3Splat: Instance-Level Open-Vocabulary 3D Scene Understanding in Gaussian Splatting
arXiv:2605.04506v2 Announce Type: replace-cross Abstract: We introduce Ilov3Splat, a novel framework for instance-level open-vocabulary 3D scene understanding built on 3D Gaussian Splatting (3D-GS). Most prior work depends on 2D rendering-based matching or point-level semantic association, which undermines cross-view consistency, lacks coherent instance-level reasoning, and limits precision in downstream 3D tasks. To address these limitations, our […]
Characteristic Root Analysis and Regularization for Linear Time Series Forecasting
arXiv:2509.23597v5 Announce Type: replace-cross Abstract: Time series forecasting remains a critical challenge across numerous domains, yet the effectiveness of complex models often varies unpredictably across datasets. Recent studies highlight the surprising competitiveness of simple linear models, suggesting that their robustness and interpretability warrant deeper theoretical investigation. This paper presents a systematic study of linear models […]
Locale-Conditioned Few-Shot Prompting Mitigates Demonstration Regurgitation in On-Device PII Substitution with Small Language Models
arXiv:2605.13538v1 Announce Type: cross Abstract: Personally Identifiable Information (PII) redaction usually replaces detected entities with placeholder tokens such as [PERSON], destroying the downstream utility of the redacted text for retrieval and Named Entity Recognition (NER) training. We propose a fully on-device pipeline that substitutes PII with consistent, type-preserving fake values: a 1.5 B mixture-of-experts token […]