Showing 1–50 of 209 results for author: Nguyen, T D

D3D-GEN: Robot-Aware Domain-Grounded Interactive 3D World Generation for Social Robotics

Authors: Anh Duc Do, Volodymyr Scherbyna, Tai Duc Nguyen, Spaarsh Thakkar, Zhengcheng Shen, Teham Buiyan, Archan Misra, Linh Kästner

Abstract: Training and validation of Embodied AI for social navigation critically depends on realistic simulation environments, yet many current approaches fail to find a balance between realism and simulability. We propose D3D-GEN, a novel world generation system that combines a domain agent with a retrieval-augmented generation (RAG) pipeline grounded in that domain. Our system enables users to rapidly ge… ▽ More Training and validation of Embodied AI for social navigation critically depends on realistic simulation environments, yet many current approaches fail to find a balance between realism and simulability. We propose D3D-GEN, a novel world generation system that combines a domain agent with a retrieval-augmented generation (RAG) pipeline grounded in that domain. Our system enables users to rapidly generate domain-grounded, fully interactive 3D worlds by automating both the collection of domain knowledge and the synthesis of realistic floorplans and object placements, without dependence on any fixed 3D model database. Given a domain description prompt, the research agent collects publicly accessible domain-specific data and constructs a persistent domain database. Using this database, our RAG pipeline generates plausible floorplans and object placements by dynamically querying a user-provided semantic database, which can be easily extended or modified. The output is a fully interactive 3D world loadable by the popular simulators Isaac Sim and Gazebo. With our approach, we have built databases for several common domains (indoor residential, hospital, office) and generated dozens of distinct, plausible simulation environments for each domain. We present D3D-GEN with a local web frontend that facilitates rapid, interactive world generation for robot simulation. △ Less

Submitted 12 August, 2026; originally announced August 2026.

Comments: 8 pages, 5 figures, and 5 tables. Accepted at the 2026 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2026)

MSC Class: 68T40 ACM Class: I.2.9; I.3.5

CoDiR: Confidence-Guided Diffusion Refinement for Semi-Supervised Histopathology Segmentation

Authors: Hoai Nhan Pham, Dang-Nguyen Bui, Le-Van Thai, Thanh-Hiep Vo, Lan Anh Dinh Thi, Tien Dat Nguyen, Duy-Dong Nguyen, Ngoc Lam Quang Bui, Tam Tran, Zhi Huang

Abstract: Semi-supervised histopathology segmentation is challenging due to scarce annotations and unreliable pseudo-labels in ambiguous gland regions. To address this problem, we propose Confidence-Guided Diffusion Refinement (CoDiR), a semi-supervised framework that combines a Mean Teacher segmentation model with diffusion-based pseudo-label refinement. Given an unlabeled image, the teacher first produces… ▽ More Semi-supervised histopathology segmentation is challenging due to scarce annotations and unreliable pseudo-labels in ambiguous gland regions. To address this problem, we propose Confidence-Guided Diffusion Refinement (CoDiR), a semi-supervised framework that combines a Mean Teacher segmentation model with diffusion-based pseudo-label refinement. Given an unlabeled image, the teacher first produces a soft prediction, and only low-confidence regions are refined by a conditional diffusion model trained to capture plausible mask structures from labeled data. The refined mask is then fused with reliable teacher predictions and used to train the student with confidence weighting and consistency regularization. On the GlaS and CRAG datasets CoDiR reaches 88.09\% and 89.83\% mDice with 10\% labeled data, and 89.19\% and 90.29\% mDice with 20\%, matching or exceeding the strongest published method on seven of the eight benchmark metrics. Ablations attribute the largest single contribution to the refinement module, which adds +6.36\% mDice over the Mean Teacher baseline. The implementation code is publicly available at: https://github.com/vongla345/codir △ Less

Submitted 12 August, 2026; originally announced August 2026.

Comments: Accepted to the MICCAI COMPAYL Workshop 2026 (11 pages, 2 figures, 6 tables)

UMSS: Towards Unsupervised Multi-modal Semantic Segmentation

Authors: Haitian Zhang, Thai Duy Nguyen, Xiangyuan Wang, Mohan Liu, Lin Wang

Abstract: Multimodal semantic segmentation (MSS) is essential for robust perception in complex environments, yet its potential remains largely untapped because of the prohibitive cost of human annotations. While unsupervised semantic segmentation (USS) has achieved strong results on a single RGB modality, its naive extension to multimodal data is often hindered by fusion degradation. This occurs because, wi… ▽ More Multimodal semantic segmentation (MSS) is essential for robust perception in complex environments, yet its potential remains largely untapped because of the prohibitive cost of human annotations. While unsupervised semantic segmentation (USS) has achieved strong results on a single RGB modality, its naive extension to multimodal data is often hindered by fusion degradation. This occurs because, without explicit supervision, existing frameworks struggle to reconcile the heterogeneous structural patterns captured by different sensors and therefore fail to effectively exploit their complementary information. In this paper, we make the first attempt to address the novel problem of Unsupervised Multimodal Semantic Segmentation (UMSS), aiming to effectively exploit complementary sensor information in a fully label free setting. To this end, we propose UniM2 (Unified Multimodal), a novel framework built on DINOv3 that transforms conventional fusion methods into consistent performance gains. Our key idea is to learn a unified latent space driven by Cross Modal Correspondence Synergy (CMCS) to extract intrinsic shared semantic cues, bypassing the need for label guided adaptive fusion. To mitigate inherent intermodal conflicts, we introduce a Cross Modal Harmonizer (CMH) that designates RGB as a stable reference, effectively suppressing inconsistent relational supervision while guiding the model to exploit complementary structural features. Extensive experimental results on NYU Depth v2 and MFNet show that UniM2 improves mIoU by 6.4% and 9.8%, respectively, demonstrating clear advantages over existing frameworks for UMSS. △ Less

Submitted 14 July, 2026; originally announced July 2026.

Trustworthy Image Authentication using Forensic Knowledge Graphs

Authors: Tai D. Nguyen, Matthew C. Stamm

Abstract: Advances in generative AI have made image falsification highly realistic, demanding trustworthy authentication systems. Existing forensic detectors can target certain forgery types but lack interpretability, while vision-language models (VLMs) provide explanations but cannot exploit forensic traces for reliable detection. We propose Forensic Knowledge Graphs (FKGs), a unified framework that integr… ▽ More Advances in generative AI have made image falsification highly realistic, demanding trustworthy authentication systems. Existing forensic detectors can target certain forgery types but lack interpretability, while vision-language models (VLMs) provide explanations but cannot exploit forensic traces for reliable detection. We propose Forensic Knowledge Graphs (FKGs), a unified framework that integrates forensic evidence extraction, structured reasoning, and human-interpretable explanation. Our FKG structure encodes forensic traces along with their causal dependencies and links to scene content. To generate accurate FKGs, we introduce a novel forensic authentication network and an Iterative Context Refinement strategy that guides VLMs to produce faithful, grounded explanations. We also present FKG-50K, a dataset of 50,000 realistic forgeries with ground-truth FKGs. Experiments demonstrate that FKG outperforms both forensic detectors and VLMs in detection, forgery identification and localization, and forensic justification. △ Less

Submitted 22 June, 2026; originally announced June 2026.

Comments: Accepted and Published at ECCV 2026

Towards Pedagogically Aligned LLM Tutors for Math Mistake Remediation

Authors: Kseniia Petukhova, Tien Dat Nguyen, Ekaterina Kochmar

Abstract: Large language models have strong potential for use in intelligent tutoring systems, but they often fail to follow effective pedagogical strategies, such as guiding students without revealing final answers. We study the application of a two-stage alignment pipeline for math mistake remediation, combining supervised fine-tuning on tutoring dialogs with Direct Preference Optimization on synthetic pr… ▽ More Large language models have strong potential for use in intelligent tutoring systems, but they often fail to follow effective pedagogical strategies, such as guiding students without revealing final answers. We study the application of a two-stage alignment pipeline for math mistake remediation, combining supervised fine-tuning on tutoring dialogs with Direct Preference Optimization on synthetic preference pairs. We construct a dataset that integrates existing tutoring corpora with synthetic data generated along pedagogical dimensions, such as scaffolding and factuality, and study different input configurations that incorporate solution correctness and gold answers. Experiments show that this approach improves both factual accuracy and pedagogical quality over base models and existing tutoring models. Human evaluation further indicates that our best model is competitive with a strong proprietary baseline, while providing additional benefits in terms of openness, transparency, and reproducibility. Our results highlight the effectiveness of preference-based pedagogical alignment, while also revealing challenges in reliably evaluating tutoring quality. △ Less

Submitted 19 June, 2026; originally announced June 2026.

Social welfare optimisation under institutional reward and punishment

Authors: Van An Nguyen, Vuong Khang Huynh, Huu Loi Bui, Hai Anh Ha, Quang Dung Le, Tan Dat Nguyen, Ngoc Ngu Nguyen, Zhao Song, Manh Hong Duong, Le Hong Trang, The Anh Han

Abstract: Institutional incentives are widely used to promote cooperation among autonomous, self-regarding agents, from human societies to multi-agent and AI systems. Existing work typically treats incentive design as a bi-objective problem: minimise institutional cost while achieving a high long-run frequency of cooperation. Whether such schemes also maximise social welfare - total population payoff net of… ▽ More Institutional incentives are widely used to promote cooperation among autonomous, self-regarding agents, from human societies to multi-agent and AI systems. Existing work typically treats incentive design as a bi-objective problem: minimise institutional cost while achieving a high long-run frequency of cooperation. Whether such schemes also maximise social welfare - total population payoff net of institutional expenditure - has remained largely unexplored. We develop a welfare-centric framework for institutional incentives in finite, well-mixed populations playing a social dilemma (Donation Game and Public Goods Game), considering both rewards for cooperators and punishments for defectors. For each mechanism, we derive explicit expressions for expected social welfare and characterise how it depends on incentive efficiency and selection intensity. Analytically, we identify parameter regimes where social welfare has a single optimal incentive level and regimes with qualitative phase transitions, in which welfare becomes non-monotonic with multiple local optima. We prove that any welfare-maximising incentive is either zero or concentrated around a simple closed-form target, and we provide an efficient algorithm to compute these optima. Comparing reward and punishment, we further derive close-formed conditions under which reward outperform punishment in terms of social welfare for any given budget. Overall, our results reveal a systematic gap between incentives optimised for cost or cooperation frequency and those that maximise welfare. △ Less

Submitted 29 May, 2026; originally announced May 2026.

Spectral Vision Transformer for Efficient Tokenization with Limited Data

Authors: Alexandra G. Roberts, Maneesh John, Jinwei Zhang, Dominick Romano, Mert Sisman, Ki Sueng Choi, Heejong Kim, Mert R. Sabuncu, Thanh D. Nguyen, Alexey V. Dimov, Pascal Spincemaille, Brian H. Kopell, Yi Wang

Abstract: We propose a novel spectral vision transformer architecture for efficient tokenization in limited data, with an emphasis on medical imaging. We outline convenient theoretical properties arising from the choice of basis including spatial invariance and optimal signal-to-noise ratio. We show reduced complexity arising from the spectral projection compared to spatial vision transformers. We show equi… ▽ More We propose a novel spectral vision transformer architecture for efficient tokenization in limited data, with an emphasis on medical imaging. We outline convenient theoretical properties arising from the choice of basis including spatial invariance and optimal signal-to-noise ratio. We show reduced complexity arising from the spectral projection compared to spatial vision transformers. We show equitable or superior performance with a reduced number of parameters as compared to a variety of models including compact and standard vision transformers, convolutional neural networks with attention, shifted window transformers, multi-layer perceptrons, and logistic regression. We include simulated, public, and clinical data in our analysis and release our code at: \verb+github.com/agr78/spectralViT+. △ Less

Submitted 12 May, 2026; originally announced May 2026.

Med-StepBench: A Hierarchical Reasoning Framework for Evaluating Hallucinations in Medical Vision-Language Models

Authors: Minh Khoi Nguyen, Dai Lam Le, Amir Reza Jafari, Tuan Dung Nguyen, Mai Hong Son, Mai Huy Thong, Quang Huy Nguyen, Thanh Trung Nguyen, Reza Farahbakhsh, Noel Crespi, Phi Le Nguyen

Abstract: Large vision-language models (VLMs) demonstrate strong performance in medical image understanding, but frequently generate clinically plausible yet incorrect statements, raising significant safety concerns. Existing medical hallucination benchmarks primarily focus on 2D imaging with one-shot diagnostic questions, offering limited insight into whether predictions are grounded in correct localizatio… ▽ More Large vision-language models (VLMs) demonstrate strong performance in medical image understanding, but frequently generate clinically plausible yet incorrect statements, raising significant safety concerns. Existing medical hallucination benchmarks primarily focus on 2D imaging with one-shot diagnostic questions, offering limited insight into whether predictions are grounded in correct localization and abnormality identification, allowing critical reasoning errors to remain hidden behind seemingly correct diagnoses. We introduce Med-StepBench, the first large-scale benchmark for step-wise hallucination detection in 3D oncological PET/CT, comprising over 12,000 images and more than 1,000,000 image-statement pairs across volumetric and multi-view 2D data, which decomposes clinical reasoning into four expert-designed diagnostic stages. Using clinician-verified annotations, we perform the first step-level evaluation of general-purpose and medical VLMs, revealing systematic failure modes obscured by aggregate accuracy metrics. Furthermore, we show that current VLMs are highly susceptible to adversarial yet clinically plausible intermediate explanations, which significantly amplify hallucinations despite contradictory visual evidence. Together, our findings highlight fundamental limitations in grounding multi-step clinical reasoning and establish Med-StepBench as a rigorous benchmark for developing safer and more reliable medical VLMs. △ Less

Submitted 11 May, 2026; originally announced May 2026.

Comments: Accepted at IJCAI-ECAI 2026

Tree transducers of linear size-to-height increase (and the additive conjunction of linear logic)

Authors: Luc Dartois, Lê Thành Dũng Nguyên, Charles Peyrat

Abstract: We investigate a natural generalization to trees of Hennie machines, a known automaton model for regular string functions. Tree-to-tree Hennie machines are tree-walking tree transducers with the ability to rewrite the node labels of their input tree, subject to a bounded visit restriction. Interestingly, they do not merely compute regular tree functions (i.e. MSO transductions), but a larger class… ▽ More We investigate a natural generalization to trees of Hennie machines, a known automaton model for regular string functions. Tree-to-tree Hennie machines are tree-walking tree transducers with the ability to rewrite the node labels of their input tree, subject to a bounded visit restriction. Interestingly, they do not merely compute regular tree functions (i.e. MSO transductions), but a larger class of functions with linear size-to-height increase (LSHI). We prove that this class sits between LSHI macro tree transducers (MTTs) and MSO set interpretations. To argue for its robustness, we show that it is closed under precomposition (resp. postcomposition) by MTTs of linear size (resp. height) increase. As a consequence, it contains the entire composition hierarchy of MTTs of linear height increase; we also prove that this composition hierarchy is strict. Finally, we give an alternative characterization of this function class based on a lambda-calculus with linear types. The key difference with similar characterizations of MSO transductions is the use of additive tuples in the encoding of output trees. Our equivalence proof, using game semantics / geometry of interaction, is heavily inspired by an analogous result on higher-order recursion schemes. △ Less

Submitted 5 May, 2026; originally announced May 2026.

PRIME: Protein Representation via Physics-Informed Multiscale Equivariant Hierarchies

Authors: Viet Thanh Duy Nguyen, John K. Johnstone, Truong-Son Hy

Abstract: Proteins are inherently multiscale physical systems whose functional properties emerge from coordinated structural organization across multiple spatial resolutions, ranging from atomic interactions to global fold topology. However, existing protein representation learning methods typically operate at a single structural level or treat different sources of structural information as parallel modalit… ▽ More Proteins are inherently multiscale physical systems whose functional properties emerge from coordinated structural organization across multiple spatial resolutions, ranging from atomic interactions to global fold topology. However, existing protein representation learning methods typically operate at a single structural level or treat different sources of structural information as parallel modalities, without explicitly modeling their hierarchical relationships. We introduce PRIME (Protein Representation via Physics-Informed Multiscale Equivariant Hierarchies), a unified framework that models proteins as a nested family of five physically grounded structural graphs spanning surface, atomic, residue, secondary-structure, and protein levels. Adjacent levels are connected through deterministic, physics-informed assignment operators, enabling bidirectional information exchange via bottom-up aggregation and top-down contextual refinement. Experiments on standard protein representation learning benchmarks demonstrate strong and competitive performance across diverse tasks, with particularly notable gains on the Fold Classification benchmark, where PRIME outperforms the strongest geometric GNN baseline by margins of 13.80 and 18.30 points on the harder Superfamily and Fold splits, and achieves a state-of-the-art accuracy of 84.10\% on Reaction Class prediction, surpassing all baseline methods, including ESM. Ablation studies confirm that each structural level contributes complementary and non-redundant information, and adaptive cross-attention analysis reveals that PRIME autonomously identifies the most task-relevant structural resolutions at prediction time. Our source code is publicly available at https://github.com/HySonLab/PRIME △ Less

Submitted 14 July, 2026; v1 submitted 2 May, 2026; originally announced May 2026.

Solving Positive Linear Programs with Differential Privacy

Authors: Alina Ene, Huy Le Nguyen, Ta Duy Nguyen, Adrian Vladu

Abstract: We study differentially private approximation algorithms for positive linear programs (LPs with nonnegative coefficients and variables), focusing on the fundamental families of packing, covering, and mixed packing-covering formulations. We focus on the high-sensitivity, constraint-private regime of Hsu-Roth-Roughgarden-Ullman (ICALP 2014), where neighboring instances may differ by an arbitrary sin… ▽ More We study differentially private approximation algorithms for positive linear programs (LPs with nonnegative coefficients and variables), focusing on the fundamental families of packing, covering, and mixed packing-covering formulations. We focus on the high-sensitivity, constraint-private regime of Hsu-Roth-Roughgarden-Ullman (ICALP 2014), where neighboring instances may differ by an arbitrary single constraint, so one cannot hope to approximately satisfy every constraint under privacy. We give private solvers that return approximate solutions while violating only a controlled number of constraints. Our algorithms improve the prior instance-dependent guarantees, and also yield new data-independent bounds that depend only on the dimension. Our techniques involve a dense multiplicative weights update method developed from a regularized dual viewpoint, which we analyze in a way that exploits structure specific to positive LPs. △ Less

Submitted 27 May, 2026; v1 submitted 29 April, 2026; originally announced April 2026.

Region-Grounded Report Generation for 3D Medical Imaging: A Fine-Grained Dataset and Graph-Enhanced Framework

Authors: Cong Huy Nguyen, Son Dinh Nguyen, Guanlin Li, Tuan Dung Nguyen, Aditya Narayan Sankaran, Mai Huy Thong, Thanh Trung Nguyen, Mai Hong Son, Reza Farahbakhsh, Phi Le Nguyen, Noel Crespi

Abstract: Automated medical report generation for 3D PET/CT imaging is fundamentally challenged by the high-dimensional nature of volumetric data and a critical scarcity of annotated datasets, particularly for low-resource languages. Current black-box methods map whole volumes to reports, ignoring the clinical workflow of analyzing localized Regions of Interest (RoIs) to derive diagnostic conclusions. In th… ▽ More Automated medical report generation for 3D PET/CT imaging is fundamentally challenged by the high-dimensional nature of volumetric data and a critical scarcity of annotated datasets, particularly for low-resource languages. Current black-box methods map whole volumes to reports, ignoring the clinical workflow of analyzing localized Regions of Interest (RoIs) to derive diagnostic conclusions. In this paper, we bridge this gap by introducing VietPET-RoI, the first large-scale 3D PET/CT dataset with fine-grained RoI annotation for a low-resource language, comprising 600 PET/CT samples and 1,960 manually annotated RoIs, paired with corresponding clinical reports. Furthermore, to demonstrate the utility of this dataset, we propose HiRRA, a novel framework that mimics the professional radiologist diagnostic workflow by employing graph-based relational modules to capture dependencies between RoI attributes. This approach shifts from global pattern matching toward localized clinical findings. Additionally, we introduce new clinical evaluation metrics, namely RoI Coverage and RoI Quality Index, that measure both RoI localization accuracy and attribute description fidelity using LLM-based extraction. Extensive evaluation demonstrates that our framework achieves SOTA performance, surpassing existing models by 19.7% in BLEU and 4.7% in ROUGE-L, while achieving a remarkable 45.8% improvement in clinical metrics, indicating enhanced clinical reliability and reduced hallucination. Our code and dataset are available on GitHub. △ Less

Submitted 15 May, 2026; v1 submitted 20 April, 2026; originally announced April 2026.

Comments: 16 pages; Accepted to appear in ACL 2026

Efficient Online 3D Multi-Camera Multi-Object Tracking and Pose Estimation

Authors: Linh Van Ma, Tran Thien Dat Nguyen, Juhua Hu, Wei Cheng, Moongu Jeon

Abstract: This paper proposes a fast and online method for jointly performing 3D multi-object tracking and pose estimation using multiple monocular cameras. Our algorithm requires only 2D bounding box and pose detections, eliminating the need for costly 3D training data or computationally expensive deep learning models. Our solution is an efficient implementation of a Bayes-optimal multi-object tracking fil… ▽ More This paper proposes a fast and online method for jointly performing 3D multi-object tracking and pose estimation using multiple monocular cameras. Our algorithm requires only 2D bounding box and pose detections, eliminating the need for costly 3D training data or computationally expensive deep learning models. Our solution is an efficient implementation of a Bayes-optimal multi-object tracking filter, enhancing computational efficiency while maintaining accuracy. We demonstrate that our algorithm is significantly faster than state-of-the-art methods without compromising accuracy, using only publicly available pre-trained 2D detection models. We also illustrate the robust performance of our algorithm in scenarios where multiple cameras are intermittently disconnected or reconnected during operation. △ Less

Submitted 12 June, 2026; v1 submitted 16 April, 2026; originally announced April 2026.

UniSemAlign: Text-Prototype Alignment with a Foundation Encoder for Semi-Supervised Histopathology Segmentation

Authors: Le-Van Thai, Tien Dat Nguyen, Hoai Nhan Pham, Lan Anh Dinh Thi, Duy-Dong Nguyen, Ngoc Lam Quang Bui

Abstract: Semi-supervised semantic segmentation in computational pathology remains challenging due to scarce pixel-level annotations and unreliable pseudo-label supervision. We propose UniSemAlign, a dual-modal semantic alignment framework that enhances visual segmentation by injecting explicit class-level structure into pixel-wise learning. Built upon a pathology-pretrained Transformer encoder, UniSemAlign… ▽ More Semi-supervised semantic segmentation in computational pathology remains challenging due to scarce pixel-level annotations and unreliable pseudo-label supervision. We propose UniSemAlign, a dual-modal semantic alignment framework that enhances visual segmentation by injecting explicit class-level structure into pixel-wise learning. Built upon a pathology-pretrained Transformer encoder, UniSemAlign introduces complementary prototype-level and text-level alignment branches in a shared embedding space, providing structured guidance that reduces class ambiguity and stabilizes pseudo-label refinement. The aligned representations are fused with visual predictions to generate more reliable supervision for unlabeled histopathology images. The framework is trained end-to-end with supervised segmentation, cross-view consistency, and cross-modal alignment objectives. Extensive experiments on the GlaS and CRAG datasets demonstrate that UniSemAlign substantially outperforms recent semi-supervised baselines under limited supervision, achieving Dice improvements of up to 2.6% on GlaS and 8.6% on CRAG with only 10% labeled data, and strong improvements at 20% supervision. Code is available at: https://github.com/thailevann/UniSemAlign △ Less

Submitted 10 April, 2026; originally announced April 2026.

Comments: Accepted at CVPR 2026 Workshop. 11 pages, 5 figures, 4 tables

Journal ref: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, 2026, pp. 6803--6813

WAND: Windowed Attention and Knowledge Distillation for Efficient Autoregressive Text-to-Speech Models

Authors: Hanna Lee, Tan Dat Nguyen, Jaehoon Kang, Kyuhong Shim

Abstract: Recent decoder-only autoregressive text-to-speech (AR-TTS) models produce high-fidelity speech, but their memory and compute costs scale quadratically with sequence length due to full self-attention. In this paper, we propose WAND, Windowed Attention and Knowledge Distillation, a framework that adapts pretrained AR-TTS models to operate with constant computational and memory complexity. WAND separ… ▽ More Recent decoder-only autoregressive text-to-speech (AR-TTS) models produce high-fidelity speech, but their memory and compute costs scale quadratically with sequence length due to full self-attention. In this paper, we propose WAND, Windowed Attention and Knowledge Distillation, a framework that adapts pretrained AR-TTS models to operate with constant computational and memory complexity. WAND separates the attention mechanism into two: persistent global attention over conditioning tokens and local sliding-window attention over generated tokens. To stabilize fine-tuning, we employ a curriculum learning strategy that progressively tightens the attention window. We further utilize knowledge distillation from a full-attention teacher to recover high-fidelity synthesis quality with high data efficiency. Evaluated on three modern AR-TTS models, WAND preserves the original quality while achieving up to 66.2% KV cache memory reduction and length-invariant, near-constant per-step latency. △ Less

Submitted 23 June, 2026; v1 submitted 17 March, 2026; originally announced April 2026.

Comments: Accepted at Interspeech 2026

Beyond the Global Scores: Fine-Grained Token Grounding as a Robust Detector of LVLM Hallucinations

Authors: Tuan Dung Nguyen, Minh Khoi Ho, Qi Chen, Yutong Xie, Nguyen Cam-Tu, Minh Khoi Nguyen, Dang Huy Pham Nguyen, Anton van den Hengel, Johan W. Verjans, Phi Le Nguyen, Vu Minh Hieu Phan

Abstract: Large vision-language models (LVLMs) achieve strong performance on visual reasoning tasks but remain highly susceptible to hallucination. Existing detection methods predominantly rely on coarse, whole-image measures of how an object token relates to the input image. This global strategy is limited: hallucinated tokens may exhibit weak but widely scattered correlations across many local regions, wh… ▽ More Large vision-language models (LVLMs) achieve strong performance on visual reasoning tasks but remain highly susceptible to hallucination. Existing detection methods predominantly rely on coarse, whole-image measures of how an object token relates to the input image. This global strategy is limited: hallucinated tokens may exhibit weak but widely scattered correlations across many local regions, which aggregate into deceptively high overall relevance, thus evading the current global hallucination detectors. We begin with a simple yet critical observation: a faithful object token must be strongly grounded in a specific image region. Building on this insight, we introduce a patch-level hallucination detection framework that examines fine-grained token-level interactions across model layers. Our analysis uncovers two characteristic signatures of hallucinated tokens: (i) they yield diffuse, non-localized attention patterns, in contrast to the compact, well-focused attention seen in faithful tokens; and (ii) they fail to exhibit meaningful semantic alignment with any visual region. Guided by these findings, we develop a lightweight and interpretable detection method that leverages patch-level statistical features, combined with hidden-layer representations. Our approach achieves up to 90% accuracy in token-level hallucination detection, demonstrating the superiority of fine-grained structural analysis for detecting hallucinations. △ Less

Submitted 6 April, 2026; originally announced April 2026.

Comments: Accepted at CVPR2026 Main Track

Transferable Multi-Bit Watermarking Across Frozen Diffusion Models via Latent Consistency Bridges

Authors: Hong-Hanh Nguyen-Le, Van-Tuan Tran, Thuc D. Nguyen, Nhien-An Le-Khac

Abstract: As generative AI advances, global governance frameworks increasingly mandate verifiable content provenance. However, existing watermarking techniques face a critical policy-to-technology disconnect: sampling-based methods require computationally prohibitive inversion, while fine-tuning approaches are tethered to specific model checkpoints, hindering standardized, cross-model oversight. To bridge t… ▽ More As generative AI advances, global governance frameworks increasingly mandate verifiable content provenance. However, existing watermarking techniques face a critical policy-to-technology disconnect: sampling-based methods require computationally prohibitive inversion, while fine-tuning approaches are tethered to specific model checkpoints, hindering standardized, cross-model oversight. To bridge this gap, we introduce DiffMark, a plug-and-play multi-bit watermarking framework. DiffMark embeds a persistent, learned perturbation into every denoising step of a frozen diffusion model, accumulating a recoverable signal in the final latent space. To enable efficient training through the frozen network, we utilize Latent Consistency Models (LCMs) as a differentiable training bridge. DiffMark achieves 64-bit extraction in a single 16.4 ms forward pass, which is a $45\times$ speed-up over inversion baselines. By enabling per-image key flexibility and cross-architecture transferability without retraining, DiffMark provides the practical, scalable technical tooling necessary to operationalize user accountability and enforce emerging AI governance mandates. △ Less

Submitted 2 June, 2026; v1 submitted 19 March, 2026; originally announced March 2026.

Comments: Accepted in Second Workshop on Technical AI Governance Research (TAIGR) @ ICML 2026

Overthinking Causes Hallucination: Tracing Confounder Propagation in Vision Language Models

Authors: Abin Shoby, Ta Duc Huy, Tuan Dung Nguyen, Minh Khoi Ho, Qi Chen, Anton van den Hengel, Phi Le Nguyen, Johan W. Verjans, Vu Minh Hieu Phan

Abstract: Vision Language models (VLMs) often hallucinate non-existent objects. Detecting hallucination is analogous to detecting deception: a single final statement is insufficient, one must examine the underlying reasoning process. Yet existing detectors rely mostly on final-layer signals. Attention-based methods assume hallucinated tokens exhibit low attention, while entropy-based ones use final-step unc… ▽ More Vision Language models (VLMs) often hallucinate non-existent objects. Detecting hallucination is analogous to detecting deception: a single final statement is insufficient, one must examine the underlying reasoning process. Yet existing detectors rely mostly on final-layer signals. Attention-based methods assume hallucinated tokens exhibit low attention, while entropy-based ones use final-step uncertainty. Our analysis reveals the opposite: hallucinated objects can exhibit peaked attention due to contextual priors; and models often express high confidence because intermediate layers have already converged to an incorrect hypothesis. We show that the key to hallucination detection lies within the model's thought process, not its final output. By probing decoder layers, we uncover a previously overlooked behavior, overthinking: models repeatedly revise object hypotheses across layers before committing to an incorrect answer. Once the model latches onto a confounded hypothesis, it can propagate through subsequent layers, ultimately causing hallucination. To capture this behavior, we introduce the Overthinking Score, a metric to measure how many competing hypotheses the model entertains and how unstable these hypotheses are across layers. This score significantly improves hallucination detection: 78.9% F1 on MSCOCO and 71.58% on AMBER. △ Less

Submitted 29 March, 2026; v1 submitted 8 March, 2026; originally announced March 2026.

Comments: CVPR2026 Findings

Adaptive Power Iteration Method for Differentially Private PCA

Authors: Ta Duy Nguyen, Alina Ene, Huy Le Nguyen

Abstract: We study $\left(ε,δ\right)$-differentially private algorithms for the problem of approximately computing the top singular vector of a matrix $A\in\mathbb{R}^{n\times d}$ where each row of $A$ is a data point in $\mathbb{R}^{d}$. Following Dwork-Talwar-Thakurta-Zhang (STOC 2014), we consider the privacy model where neighboring inputs differ by one single row. We give a novel algorithm that achieves… ▽ More We study $\left(ε,δ\right)$-differentially private algorithms for the problem of approximately computing the top singular vector of a matrix $A\in\mathbb{R}^{n\times d}$ where each row of $A$ is a data point in $\mathbb{R}^{d}$. Following Dwork-Talwar-Thakurta-Zhang (STOC 2014), we consider the privacy model where neighboring inputs differ by one single row. We give a novel algorithm that achieves beyond-worst-case guarantees for input matrices with low coherence, which is a structural property of matrices in many applications, including but not limited to i.i.d. data. Our algorithm contributes to the extensive literature on private power iteration methods, where we introduce a new filtering technique which adapts to this coherence parameter. Our work departs from and complements the work by Hardt-Roth (STOC 2013) which achieves beyond-worst-case guarantees for the more restrictive privacy model where neighboring inputs differ in one single entry by at most 1. △ Less

Submitted 19 May, 2026; v1 submitted 11 February, 2026; originally announced February 2026.

Multimodal Machine Learning for Soft High-k Elastomers under Data Scarcity

Authors: Brijesh FNU, Viet Thanh Duy Nguyen, Ashima Sharma, Md Harun Rashid Molla, Chengyi Xu, Truong-Son Hy

Abstract: Dielectric materials are critical building blocks for modern electronics such as sensors, actuators, and transistors. With rapid advances in soft and stretchable electronics for emerging human- and robot-interfacing applications, there is a growing need for high-performance dielectric elastomers. However, developing soft elastomers that simultaneously exhibit high dielectric constants (k) and low… ▽ More Dielectric materials are critical building blocks for modern electronics such as sensors, actuators, and transistors. With rapid advances in soft and stretchable electronics for emerging human- and robot-interfacing applications, there is a growing need for high-performance dielectric elastomers. However, developing soft elastomers that simultaneously exhibit high dielectric constants (k) and low Young's moduli (E) remains a major challenge. Although individual elastomer designs have been reported, structured datasets that systematically integrate molecular sequence, dielectric, and mechanical properties are largely unavailable. To address this gap, we curate a compact, high-quality dataset of acrylate-based dielectric elastomers by aggregating experimental results from the past decade. Building on this dataset, we propose a multimodal learning framework leveraging large-scale pretrained polymer representations. These pretrained embeddings transfer chemical and structural knowledge from vast polymer corpora, enabling accurate few-shot prediction of dielectric and mechanical properties and accelerating data-efficient discovery of soft high-$k$ dielectric elastomers. Our data and implementation are publicly available at: https://github.com/HySonLab/Polymers △ Less

Submitted 18 March, 2026; v1 submitted 25 January, 2026; originally announced January 2026.

Representation Learning with Semantic-aware Instance and Sparse Token Alignments

Authors: Phuoc-Nguyen Bui, Toan Duc Nguyen, Junghyun Bum, Duc-Tai Le, Hyunseung Choo

Abstract: Medical contrastive vision-language pre-training (VLP) has demonstrated significant potential in improving performance on downstream tasks. Traditional approaches typically employ contrastive learning, treating paired image-report samples as positives and unpaired ones as negatives. However, in medical datasets, there can be substantial similarities between images or reports from different patient… ▽ More Medical contrastive vision-language pre-training (VLP) has demonstrated significant potential in improving performance on downstream tasks. Traditional approaches typically employ contrastive learning, treating paired image-report samples as positives and unpaired ones as negatives. However, in medical datasets, there can be substantial similarities between images or reports from different patients. Rigidly treating all unpaired samples as negatives, can disrupt the underlying semantic structure and negatively impact the quality of the learned representations. In this paper, we propose a multi-level alignment framework, Representation Learning with Semantic-aware Instance and Sparse Token Alignments (SISTA) by exploiting the semantic correspondence between medical image and radiology reports at two levels, i.e., image-report and patch-word levels. Specifically, we improve the conventional contrastive learning by incorporating inter-report similarity to eliminate the false negatives and introduce a method to effectively align image patches with relevant word tokens. Experimental results demonstrate the effectiveness of the proposed framework in improving transfer performance across different datasets on three downstream tasks: image classification, image segmentation, and object detection. Notably, our framework achieves significant improvements in fine-grained tasks even with limited labeled data. Codes and pre-trained models will be made available. △ Less

Submitted 1 April, 2026; v1 submitted 12 January, 2026; originally announced January 2026.

Comments: Accepted to ICPR 2026

Multi-Agent Deep Reinforcement Learning for Collaborative UAV Relay Networks under Jamming Atatcks

Authors: Thai Duong Nguyen, Ngoc-Tan Nguyen, Thanh-Dao Nguyen, Nguyen Van Huynh, Dinh-Hieu Tran, Symeon Chatzinotas

Abstract: The deployment of Unmanned Aerial Vehicle (UAV) swarms as dynamic communication relays is critical for next-generation tactical networks. However, operating in contested environments requires solving a complex trade-off, including maximizing system throughput while ensuring collision avoidance and resilience against adversarial jamming. Existing heuristic-based approaches often struggle to find ef… ▽ More The deployment of Unmanned Aerial Vehicle (UAV) swarms as dynamic communication relays is critical for next-generation tactical networks. However, operating in contested environments requires solving a complex trade-off, including maximizing system throughput while ensuring collision avoidance and resilience against adversarial jamming. Existing heuristic-based approaches often struggle to find effective solutions due to the dynamic and multi-objective nature of this problem. This paper formulates this challenge as a cooperative Multi-Agent Reinforcement Learning (MARL) problem, solved using the Centralized Training with Decentralized Execution (CTDE) framework. Our approach employs a centralized critic that uses global state information to guide decentralized actors which operate using only local observations. Simulation results show that our proposed framework significantly outperforms heuristic baselines, increasing the total system throughput by approximately 50% while simultaneously achieving a near-zero collision rate. A key finding is that the agents develop an emergent anti-jamming strategy without explicit programming. They learn to intelligently position themselves to balance the trade-off between mitigating interference from jammers and maintaining effective communication links with ground users. △ Less

Submitted 9 December, 2025; originally announced December 2025.

Comments: IEEE ICC 2026

Social welfare optimisation in well-mixed and structured populations

Authors: Van An Nguyen, Vuong Khang Huynh, Ho Nam Duong, Huu Loi Bui, Hai Anh Ha, Quang Dung Le, Le Quoc Dung Ngo, Tan Dat Nguyen, Ngoc Ngu Nguyen, Hoai Thuong Nguyen, Zhao Song, Le Hong Trang, The Anh Han

Abstract: Research on promoting cooperation among autonomous, self-regarding agents has often focused on the bi-objective optimisation problem: minimising the total incentive cost while maximising the frequency of cooperation. However, the optimal value of social welfare under such constraints remains largely unexplored. In this work, we hypothesise that achieving maximal social welfare is not guaranteed at… ▽ More Research on promoting cooperation among autonomous, self-regarding agents has often focused on the bi-objective optimisation problem: minimising the total incentive cost while maximising the frequency of cooperation. However, the optimal value of social welfare under such constraints remains largely unexplored. In this work, we hypothesise that achieving maximal social welfare is not guaranteed at the minimal incentive cost required to drive agents to a desired cooperative state. To address this gap, we adopt to a single-objective approach focused on maximising social welfare, building upon foundational evolutionary game theory models that examined cost efficiency in finite populations, in both well-mixed and structured population settings. Our analytical model and agent-based simulations show how different interference strategies, including rewarding local versus global behavioural patterns, affect social welfare and dynamics of cooperation. Our results reveal a significant gap in the per-individual incentive cost between optimising for pure cost efficiency or cooperation frequency and optimising for maximal social welfare. Overall, our findings indicate that incentive design, policy, and benchmarking in multi-agent systems and human societies should prioritise welfare-centric objectives over proxy targets of cost or cooperation frequency. △ Less

Submitted 14 December, 2025; v1 submitted 8 December, 2025; originally announced December 2025.

A finer reparameterisation theorem for MSO and FO queries on strings

Authors: Lê Thành Dũng Nguyên, Paweł Parys

Abstract: We show a theorem on monadic second-order k-ary queries on finite words. It may be illustrated by the following example: if the number of results of a query on binary strings is O(number of 0s $\times$ number of 1s), then each result can be MSO-definably identified from a 0-position, a 1-position and some finite data. Our proofs also handle the case of first-order logic / aperiodic monoids. Thus… ▽ More We show a theorem on monadic second-order k-ary queries on finite words. It may be illustrated by the following example: if the number of results of a query on binary strings is O(number of 0s $\times$ number of 1s), then each result can be MSO-definably identified from a 0-position, a 1-position and some finite data. Our proofs also handle the case of first-order logic / aperiodic monoids. Thus we can state and prove the folklore theorem that dimension minimisation holds for first-order string-to-string interpretations. △ Less

Submitted 22 May, 2026; v1 submitted 6 December, 2025; originally announced December 2025.

Comments: 5 pages; not submitted to a journal yet, some details need to be fleshed out. New in v2: proof of counterexample, via N-rational series; added recent references

Reinforcing Trustworthiness in Multimodal Emotional Support Systems

Authors: Huy M. Le, Dat Tien Nguyen, Ngan T. T. Vo, Tuan D. Q. Nguyen, Nguyen Binh Le, Duy Minh Ho Nguyen, Daniel Sonntag, Lizi Liao, Binh T. Nguyen

Abstract: In today's world, emotional support is increasingly essential, yet it remains challenging for both those seeking help and those offering it. Multimodal approaches to emotional support show great promise by integrating diverse data sources to provide empathetic, contextually relevant responses, fostering more effective interactions. However, current methods have notable limitations, often relying s… ▽ More In today's world, emotional support is increasingly essential, yet it remains challenging for both those seeking help and those offering it. Multimodal approaches to emotional support show great promise by integrating diverse data sources to provide empathetic, contextually relevant responses, fostering more effective interactions. However, current methods have notable limitations, often relying solely on text or converting other data types into text, or providing emotion recognition only, thus overlooking the full potential of multimodal inputs. Moreover, many studies prioritize response generation without accurately identifying critical emotional support elements or ensuring the reliability of outputs. To overcome these issues, we introduce \textsc{ MultiMood}, a new framework that (i) leverages multimodal embeddings from video, audio, and text to predict emotional components and to produce responses responses aligned with professional therapeutic standards. To improve trustworthiness, we (ii) incorporate novel psychological criteria and apply Reinforcement Learning (RL) to optimize large language models (LLMs) for consistent adherence to these standards. We also (iii) analyze several advanced LLMs to assess their multimodal emotional support capabilities. Experimental results show that MultiMood achieves state-of-the-art on MESC and DFEW datasets while RL-driven trustworthiness improvements are validated through human and LLM evaluations, demonstrating its superior capability in applying a multimodal framework in this domain. △ Less

Submitted 17 November, 2025; v1 submitted 13 November, 2025; originally announced November 2025.

Quantum Reinforcement Learning for 6G and Beyond Wireless Networks

Authors: Dinh-Hieu Tran, Thai Duong Nguyen, Thanh-Dao Nguyen, Ngoc-Tan Nguyen, Van Nhan Vo, Hung Tran, Mouhamad Chehaitly, Yan Kyaw Tun, Cedomir Stefanovic, Tu Ho Dac, Eva Lagunas, Symeon Chatzinotas, Nguyen Van Huynh

Abstract: While 5G is being deployed worldwide, 6G is receiving increasing attention from researchers to meet the growing demand for higher data rates, lower latency, higher density, and seamless communications worldwide. To meet the stringent requirements of 6G wireless communications networks, AI-integrated communications have become an indispensable part of supporting 6G systems with intelligence, automa… ▽ More While 5G is being deployed worldwide, 6G is receiving increasing attention from researchers to meet the growing demand for higher data rates, lower latency, higher density, and seamless communications worldwide. To meet the stringent requirements of 6G wireless communications networks, AI-integrated communications have become an indispensable part of supporting 6G systems with intelligence, automation, and big data training capabilities. However, traditional artificial intelligence (AI) systems are difficult to meet the stringent latency and high throughput requirements of 6G with limited resources. In this article, we summarize, analyze, discuss the potential, and benefits of Quantum Reinforcement Learning (QRL) in 6G. As an example, we show the superiority of QRL in dynamic spectrum access compared to the conventional Deep Reinforcement Learning (DRL) approach. In addition, we provide an overview of what DRL has accomplished in 6G and its challenges and limitations. From there, we introduce QRL and potential research directions that should continue to be of interest in 6G. To the best of our knowledge, this is the first review and vision article on QRL for 6G wireless communication networks. △ Less

Submitted 2 November, 2025; originally announced November 2025.

Quasi-Self-Concordant Optimization with Lewis Weights

Authors: Alina Ene, Ta Duy Nguyen, Adrian Vladu

Abstract: In this paper, we study the problem $\min_{x\in \mathbb{R}^{d},Nx=v}\sum_{i=1}^{n}f((Ax-b)_{i})$ for a quasi-self-concordant function $f:\mathbb{R}\to\mathbb{R}$, where $A,N$ are $n\times d$ and $m\times d$ matrices, $b,v$ are vectors of length $n$ and $m$ with $n\ge d.$ We show an algorithm based on a trust-region method with an oracle that can be implemented using $\widetilde{O}(d^{1/3})$ linear… ▽ More In this paper, we study the problem $\min_{x\in \mathbb{R}^{d},Nx=v}\sum_{i=1}^{n}f((Ax-b)_{i})$ for a quasi-self-concordant function $f:\mathbb{R}\to\mathbb{R}$, where $A,N$ are $n\times d$ and $m\times d$ matrices, $b,v$ are vectors of length $n$ and $m$ with $n\ge d.$ We show an algorithm based on a trust-region method with an oracle that can be implemented using $\widetilde{O}(d^{1/3})$ linear system solves, improving the $\widetilde{O}(n^{1/3})$ oracle by {[}Adil-Bullins-Sachdeva, NeurIPS 2021{]}. Our implementation of the oracle relies on solving the overdetermined $\ell_{\infty}$-regression problem $\min_{x\in\mathbb{R}^{d},Nx=v}\|Ax-b\|_{\infty}$. We provide an algorithm that finds a $(1+ε)$-approximate solution to this problem using $O((d^{1/3}/ε+1/ε^{2})\log(n/ε))$ linear system solves. This algorithm leverages $\ell_{\infty}$ Lewis weight overestimates and achieves this iteration complexity via a simple lightweight IRLS approach, inspired by the work of {[}Ene-Vladu, ICML 2019{]}. Experimentally, we demonstrate that our algorithm significantly improves the runtime of the standard CVX solver. △ Less

Submitted 24 October, 2025; originally announced October 2025.

Unmasking Puppeteers: Leveraging Biometric Leakage to Expose Impersonation in AI-Based Videoconferencing

Authors: Danial Samadi Vahdati, Tai Duc Nguyen, Ekta Prashnani, Koki Nagano, David Luebke, Orazio Gallo, Matthew Stamm

Abstract: AI-based talking-head videoconferencing systems reduce bandwidth by sending a compact pose-expression latent and re-synthesizing RGB at the receiver, but this latent can be puppeteered, letting an attacker hijack a victim's likeness in real time. Because every frame is synthetic, deepfake and synthetic video detectors fail outright. To address this security problem, we exploit a key observation: t… ▽ More AI-based talking-head videoconferencing systems reduce bandwidth by sending a compact pose-expression latent and re-synthesizing RGB at the receiver, but this latent can be puppeteered, letting an attacker hijack a victim's likeness in real time. Because every frame is synthetic, deepfake and synthetic video detectors fail outright. To address this security problem, we exploit a key observation: the pose-expression latent inherently contains biometric information of the driving identity. Therefore, we introduce the first biometric leakage defense without ever looking at the reconstructed RGB video: a pose-conditioned, large-margin contrastive encoder that isolates persistent identity cues inside the transmitted latent while cancelling transient pose and expression. A simple cosine test on this disentangled embedding flags illicit identity swaps as the video is rendered. Our experiments on multiple talking-head generation models show that our method consistently outperforms existing puppeteering defenses, operates in real-time, and shows strong generalization to out-of-distribution scenarios. △ Less

Submitted 12 May, 2026; v1 submitted 3 October, 2025; originally announced October 2025.

Improved $\ell_{p}$ Regression via Iteratively Reweighted Least Squares

Authors: Alina Ene, Ta Duy Nguyen, Adrian Vladu

Abstract: We introduce fast algorithms for solving $\ell_{p}$ regression problems using the iteratively reweighted least squares (IRLS) method. Our approach achieves state-of-the-art iteration complexity, outperforming the IRLS algorithm by Adil-Peng-Sachdeva (NeurIPS 2019) and matching the theoretical bounds established by the complex algorithm of Adil-Kyng-Peng-Sachdeva (SODA 2019, J. ACM 2024) via a simp… ▽ More We introduce fast algorithms for solving $\ell_{p}$ regression problems using the iteratively reweighted least squares (IRLS) method. Our approach achieves state-of-the-art iteration complexity, outperforming the IRLS algorithm by Adil-Peng-Sachdeva (NeurIPS 2019) and matching the theoretical bounds established by the complex algorithm of Adil-Kyng-Peng-Sachdeva (SODA 2019, J. ACM 2024) via a simpler lightweight iterative scheme. This bridges the existing gap between theoretical and practical algorithms for $\ell_{p}$ regression. Our algorithms depart from prior approaches, using a primal-dual framework, in which the update rule can be naturally derived from an invariant maintained for the dual objective. Empirically, we show that our algorithms significantly outperform both the IRLS algorithm by Adil-Peng-Sachdeva and MATLAB/CVX implementations. △ Less

Submitted 2 October, 2025; originally announced October 2025.

MoVa: Towards Generalizable Classification of Human Morals and Values

Authors: Ziyu Chen, Junfei Sun, Chenxi Li, Tuan Dung Nguyen, Jing Yao, Xiaoyuan Yi, Xing Xie, Chenhao Tan, Lexing Xie

Abstract: Identifying human morals and values embedded in language is essential to empirical studies of communication. However, researchers often face substantial difficulty navigating the diversity of theoretical frameworks and data available for their analysis. Here, we contribute MoVa, a well-documented suite of resources for generalizable classification of human morals and values, consisting of (1) 16 l… ▽ More Identifying human morals and values embedded in language is essential to empirical studies of communication. However, researchers often face substantial difficulty navigating the diversity of theoretical frameworks and data available for their analysis. Here, we contribute MoVa, a well-documented suite of resources for generalizable classification of human morals and values, consisting of (1) 16 labeled datasets and benchmarking results from four theoretically-grounded frameworks; (2) a lightweight LLM prompting strategy that outperforms fine-tuned models across multiple domains and frameworks; and (3) a new application that helps evaluate psychological surveys. In practice, we specifically recommend a classification strategy, all@once, that scores all related concepts simultaneously, resembling the well-known multi-label classifier chain. The data and methods in MoVa can facilitate many fine-grained interpretations of human and machine communication, with potential implications for the alignment of machine behavior. △ Less

Submitted 28 September, 2025; originally announced September 2025.

Comments: 9 pages, 10 figures and tables, EMNLP 2025 main conference

SPADE: Structured Pruning and Adaptive Distillation for Efficient LLM-TTS

Authors: Tan Dat Nguyen, Jaehun Kim, Ji-Hoon Kim, Shukjae Choi, Youshin Lim, Joon Son Chung

Abstract: The goal of this paper is to introduce SPADE, a framework for Structured Pruning and Adaptive Distillation for Efficient Large Language Model-based text-to-speech (LLM-TTS). Recent LLM-TTS systems achieve strong controllability and zero-shot generalization, but their large parameter counts and high latency limit real-world deployment. SPADE addresses this by combining (i) a pruning step guided by… ▽ More The goal of this paper is to introduce SPADE, a framework for Structured Pruning and Adaptive Distillation for Efficient Large Language Model-based text-to-speech (LLM-TTS). Recent LLM-TTS systems achieve strong controllability and zero-shot generalization, but their large parameter counts and high latency limit real-world deployment. SPADE addresses this by combining (i) a pruning step guided by a word-error-rate-based layer importance index to remove non-essential Transformer layers, with (ii) multi-level knowledge distillation to restore autoregressive coherence. On zero-shot benchmarks, SPADE preserves near-parity perceptual quality while halving Transformer depth, reducing VRAM usage by up to 20%, and achieving up to 1.7x faster real-time factor with less than 5% of the original training data. These results show that compact LLM-TTS models can maintain naturalness and speaker similarity while enabling practical real-time speech generation. Audio samples are available at https://mm.kaist.ac.kr/projects/SPADE/. △ Less

Submitted 29 January, 2026; v1 submitted 25 September, 2025; originally announced September 2025.

Comments: ICASSP 2026

MAGE: A Coarse-to-Fine Speech Enhancer with Masked Generative Model

Authors: The Hieu Pham, Tan Dat Nguyen, Phuong Thanh Tran, Joon Son Chung, Duc Dung Nguyen

Abstract: Speech enhancement remains challenging due to the trade-off between efficiency and perceptual quality. In this paper, we introduce MAGE, a Masked Audio Generative Enhancer that advances generative speech enhancement through a compact and robust design. Unlike prior masked generative models with random masking, MAGE employs a scarcity-aware coarse-to-fine masking strategy that prioritizes frequent… ▽ More Speech enhancement remains challenging due to the trade-off between efficiency and perceptual quality. In this paper, we introduce MAGE, a Masked Audio Generative Enhancer that advances generative speech enhancement through a compact and robust design. Unlike prior masked generative models with random masking, MAGE employs a scarcity-aware coarse-to-fine masking strategy that prioritizes frequent tokens in early steps and rare tokens in later refinements, improving efficiency and generalization. We also propose a lightweight corrector module that further stabilizes inference by detecting low-confidence predictions and re-masking them for refinement. Built on BigCodec and finetuned from Qwen2.5-0.5B, MAGE is reduced to 200M parameters through selective layer retention. Experiments on DNS Challenge and noisy LibriSpeech show that MAGE achieves state-of-the-art perceptual quality and significantly reduces word error rate for downstream recognition, outperforming larger baselines. Audio examples are available at https://hieugiaosu.github.io/MAGE/. △ Less

Submitted 13 March, 2026; v1 submitted 24 September, 2025; originally announced September 2025.

Comments: ICASSP 2026

LINKER: Learning Interactions Between Functional Groups and Residues With Chemical Knowledge-Enhanced Reasoning and Explainability

Authors: Phuc Pham, Viet Thanh Duy Nguyen, Truong-Son Hy

Abstract: Accurate identification of interactions between protein residues and ligand functional groups is essential to understand molecular recognition and guide rational drug design. Existing deep learning approaches for protein-ligand interpretability often rely on 3D structural input or use distance-based contact labels, limiting both their applicability and biological relevance. We introduce LINKER, th… ▽ More Accurate identification of interactions between protein residues and ligand functional groups is essential to understand molecular recognition and guide rational drug design. Existing deep learning approaches for protein-ligand interpretability often rely on 3D structural input or use distance-based contact labels, limiting both their applicability and biological relevance. We introduce LINKER, the first sequence-based model to predict residue-functional group interactions in terms of biologically defined interaction types, using only protein sequences and the ligand SMILES as input. LINKER is trained with structure-supervised attention, where interaction labels are derived from 3D protein-ligand complexes via functional group-based motif extraction. By abstracting ligand structures into functional groups, the model focuses on chemically meaningful substructures while predicting interaction types rather than mere spatial proximity. Crucially, LINKER requires only sequence-level input at inference time, enabling large-scale application in settings where structural data is unavailable. Experiments on the LP-PDBBind benchmark demonstrate that structure-informed supervision over functional group abstractions yields interaction predictions closely aligned with ground-truth biochemical annotations. △ Less

Submitted 3 September, 2025; originally announced September 2025.

SimInterview: Transforming Business Education through Large Language Model-Based Simulated Multilingual Interview Training System

Authors: Truong Thanh Hung Nguyen, Tran Diem Quynh Nguyen, Hoang Loc Cao, Thi Cam Thanh Tran, Thi Cam Mai Truong, Hung Cao

Abstract: Business interview preparation demands both solid theoretical grounding and refined soft skills, yet conventional classroom methods rarely deliver the individualized, culturally aware practice employers currently expect. This paper introduces SimInterview, a large language model (LLM)-based simulated multilingual interview training system designed for business professionals entering the AI-transfo… ▽ More Business interview preparation demands both solid theoretical grounding and refined soft skills, yet conventional classroom methods rarely deliver the individualized, culturally aware practice employers currently expect. This paper introduces SimInterview, a large language model (LLM)-based simulated multilingual interview training system designed for business professionals entering the AI-transformed labor market. Our system leverages an LLM agent and synthetic AI technologies to create realistic virtual recruiters capable of conducting personalized, real-time conversational interviews. The framework dynamically adapts interview scenarios using retrieval-augmented generation (RAG) to match individual resumes with specific job requirements across multiple languages. Built on LLMs (OpenAI o3, Llama 4 Maverick, Gemma 3), integrated with Whisper speech recognition, GPT-SoVITS voice synthesis, Ditto diffusion-based talking head generation model, and ChromaDB vector databases, our system significantly improves interview readiness across English and Japanese markets. Experiments with university-level candidates show that the system consistently aligns its assessments with job requirements, faithfully preserves resume content, and earns high satisfaction ratings, with the lightweight Gemma 3 model producing the most engaging conversations. Qualitative findings revealed that the standardized Japanese resume format improved document retrieval while diverse English resumes introduced additional variability, and they highlighted how cultural norms shape follow-up questioning strategies. Finally, we also outlined a contestable AI design that can explain, detect bias, and preserve human-in-the-loop to meet emerging regulatory expectations. △ Less

Submitted 15 August, 2025; originally announced August 2025.

Comments: Published as a conference paper at ICEFM 2025

Humanoid Robot Whole-body Geometric Calibration with Embedded Sensors and a Single Plane

Authors: Thanh D V Nguyen, Vincent Bonnet, Pierre Fernbach, David Daney, Florent Lamiraux

Abstract: Whole-body geometric calibration of humanoid robots using classical robot calibration methods is a timeconsuming and experimentally burdensome task. However, despite its significance for accurate control and simulation, it is often overlooked in the humanoid robotics community. To address this issue, we propose a novel practical method that utilizes a single plane, embedded force sensors, and an a… ▽ More Whole-body geometric calibration of humanoid robots using classical robot calibration methods is a timeconsuming and experimentally burdensome task. However, despite its significance for accurate control and simulation, it is often overlooked in the humanoid robotics community. To address this issue, we propose a novel practical method that utilizes a single plane, embedded force sensors, and an admittance controller to calibrate the whole-body kinematics of humanoids without requiring manual intervention. Given the complexity of humanoid robots, it is crucial to generate and determine a minimal set of optimal calibration postures. To do so, we propose a new algorithm called IROC (Information Ranking algorithm for selecting Optimal Calibration postures). IROC requires a pool of feasible candidate postures to build a normalized weighted information matrix for each posture. Then, contrary to other algorithms from the literature, IROC will determine the minimal number of optimal postures that are to be played onto a robot for its calibration. Both IROC and the single-plane calibration method were experimentally validated on a TALOS humanoid robot. The total whole-body kinematics chain was calibrated using solely 31 optimal postures with 3-point contacts on a table by the robot gripper. In a cross-validation experiment, the average root-mean-square (RMS) error was reduced by a factor of 2.3 compared to the manufacturer's model. △ Less

Submitted 22 July, 2025; originally announced July 2025.

Solving Linear Programs with Differential Privacy

Authors: Alina Ene, Huy Le Nguyen, Ta Duy Nguyen, Adrian Vladu

Abstract: We study the problem of solving linear programs of the form $Ax\le b$, $x\ge0$ with differential privacy. For homogeneous LPs $Ax\ge0$, we give an efficient $(ε,δ)$-differentially private algorithm which with probability at least $1-β$ finds in polynomial time a solution that satisfies all but $O(\frac{d^{2}}ε\log^{2}\frac{d}{δβ}\sqrt{\log\frac{1}{ρ_{0}}})$ constraints, for problems with margin… ▽ More We study the problem of solving linear programs of the form $Ax\le b$, $x\ge0$ with differential privacy. For homogeneous LPs $Ax\ge0$, we give an efficient $(ε,δ)$-differentially private algorithm which with probability at least $1-β$ finds in polynomial time a solution that satisfies all but $O(\frac{d^{2}}ε\log^{2}\frac{d}{δβ}\sqrt{\log\frac{1}{ρ_{0}}})$ constraints, for problems with margin $ρ_{0}>0$. This improves the bound of $O(\frac{d^{5}}ε\log^{1.5}\frac{1}{ρ_{0}}\mathrm{poly}\log(d,\frac{1}δ,\frac{1}β))$ by [Kaplan-Mansour-Moran-Stemmer-Tur, STOC '25]. For general LPs $Ax\le b$, $x\ge0$ with potentially zero margin, we give an efficient $(ε,δ)$-differentially private algorithm that w.h.p drops $O(\frac{d^{4}}ε\log^{2.5}\frac{d}δ\sqrt{\log dU})$ constraints, where $U$ is an upper bound for the entries of $A$ and $b$ in absolute value. This improves the result by Kaplan et al. by at least a factor of $d^{5}$. Our techniques build upon privatizing a rescaling perceptron algorithm by [Hoberg-Rothvoss, IPCO '17] and a more refined iterative procedure for identifying equality constraints by Kaplan et al. △ Less

Submitted 14 July, 2025; originally announced July 2025.

Enhancing Retrieval Augmented Generation with Hierarchical Text Segmentation Chunking

Authors: Hai Toan Nguyen, Tien Dat Nguyen, Viet Ha Nguyen

Abstract: Retrieval-Augmented Generation (RAG) systems commonly use chunking strategies for retrieval, which enhance large language models (LLMs) by enabling them to access external knowledge, ensuring that the retrieved information is up-to-date and domain-specific. However, traditional methods often fail to create chunks that capture sufficient semantic meaning, as they do not account for the underlying t… ▽ More Retrieval-Augmented Generation (RAG) systems commonly use chunking strategies for retrieval, which enhance large language models (LLMs) by enabling them to access external knowledge, ensuring that the retrieved information is up-to-date and domain-specific. However, traditional methods often fail to create chunks that capture sufficient semantic meaning, as they do not account for the underlying textual structure. This paper proposes a novel framework that enhances RAG by integrating hierarchical text segmentation and clustering to generate more meaningful and semantically coherent chunks. During inference, the framework retrieves information by leveraging both segment-level and cluster-level vector representations, thereby increasing the likelihood of retrieving more precise and contextually relevant information. Evaluations on the NarrativeQA, QuALITY, and QASPER datasets indicate that the proposed method achieved improved results compared to traditional chunking techniques. △ Less

Submitted 14 July, 2025; originally announced July 2025.

ClozeMath: Improving Mathematical Reasoning in Language Models by Learning to Fill Equations

Authors: Quang Hieu Pham, Thuy Duong Nguyen, Tung Pham, Anh Tuan Luu, Dat Quoc Nguyen

Abstract: The capabilities of large language models (LLMs) have been enhanced by training on data that reflects human thought processes, such as the Chain-of-Thought format. However, evidence suggests that the conventional scheme of next-word prediction may not fully capture how humans learn to think. Inspired by how humans generalize mathematical reasoning, we propose a new approach named ClozeMath to fine… ▽ More The capabilities of large language models (LLMs) have been enhanced by training on data that reflects human thought processes, such as the Chain-of-Thought format. However, evidence suggests that the conventional scheme of next-word prediction may not fully capture how humans learn to think. Inspired by how humans generalize mathematical reasoning, we propose a new approach named ClozeMath to fine-tune LLMs for mathematical reasoning. Our ClozeMath involves a text-infilling task that predicts masked equations from a given solution, analogous to cloze exercises used in human learning. Experiments on GSM8K, MATH, and GSM-Symbolic show that ClozeMath surpasses the strong baseline Masked Thought in performance and robustness, with two test-time scaling decoding algorithms, Beam Search and Chain-of-Thought decoding. Additionally, we conduct an ablation study to analyze the effects of various architectural and implementation choices on our approach. △ Less

Submitted 4 June, 2025; originally announced June 2025.

Comments: Accepted to ACL 2025 Findings

Synthetic Generation and Latent Projection Denoising of Rim Lesions in Multiple Sclerosis

Authors: Alexandra G. Roberts, Ha M. Luu, Mert Şişman, Alexey V. Dimov, Ceren Tozlu, Ilhami Kovanlikaya, Susan A. Gauthier, Thanh D. Nguyen, Yi Wang

Abstract: Quantitative susceptibility maps from magnetic resonance images can provide both prognostic and diagnostic information in multiple sclerosis, a neurodegenerative disease characterized by the formation of lesions in white matter brain tissue. In particular, susceptibility maps provide adequate contrast to distinguish between "rim" lesions, surrounded by deposited paramagnetic iron, and "non-rim" le… ▽ More Quantitative susceptibility maps from magnetic resonance images can provide both prognostic and diagnostic information in multiple sclerosis, a neurodegenerative disease characterized by the formation of lesions in white matter brain tissue. In particular, susceptibility maps provide adequate contrast to distinguish between "rim" lesions, surrounded by deposited paramagnetic iron, and "non-rim" lesion types. These paramagnetic rim lesions (PRLs) are an emerging biomarker in multiple sclerosis. Much effort has been devoted to both detection and segmentation of such lesions to monitor longitudinal change. As paramagnetic rim lesions are rare, addressing this problem requires confronting the class imbalance between rim and non-rim lesions. We produce synthetic quantitative susceptibility maps of paramagnetic rim lesions and show that inclusion of such synthetic data improves classifier performance and provide a multi-channel extension to generate accompanying contrasts and probabilistic segmentation maps. We exploit the projection capability of our trained generative network to demonstrate a novel denoising approach that allows us to train on ambiguous rim cases and substantially increase the minority class. We show that both synthetic lesion synthesis and our proposed rim lesion label denoising method best approximate the unseen rim lesion distribution and improve detection in a clinically interpretable manner. We release our code and generated data at https://github.com/agr78/PRLx-GAN upon publication. △ Less

Submitted 29 May, 2025; originally announced May 2025.

Comments: Accepted full paper in Synthetic Data @ CVPR 2025 12 pages, 10 figures

AstroMLab 4: Benchmark-Topping Performance in Astronomy Q&A with a 70B-Parameter Domain-Specialized Reasoning Model

Authors: Tijmen de Haan, Yuan-Sen Ting, Tirthankar Ghosal, Tuan Dung Nguyen, Alberto Accomazzi, Emily Herron, Vanessa Lama, Rui Pan, Azton Wells, Nesar Ramachandra

Abstract: General-purpose large language models (LLMs), despite their broad capabilities, often struggle with specialized domain knowledge. This gap hinders their deployment as reliable research agents in demanding fields such as astronomy. Building on our prior work with AstroSage-Llama-3.1-8B, this study introduces AstroSage-Llama-3.1-70B, a 70-billion parameter domain-specialized natural-language AI assi… ▽ More General-purpose large language models (LLMs), despite their broad capabilities, often struggle with specialized domain knowledge. This gap hinders their deployment as reliable research agents in demanding fields such as astronomy. Building on our prior work with AstroSage-Llama-3.1-8B, this study introduces AstroSage-Llama-3.1-70B, a 70-billion parameter domain-specialized natural-language AI assistant. It is designed for research and education across astronomy, astrophysics, space science, astroparticle physics, cosmology, and astronomical instrumentation. Developed from the Meta-Llama-3.1-70B foundation, AstroSage-Llama-3.1-70B underwent extensive continued pre-training (CPT) on a vast corpus of astronomical literature, followed by supervised fine-tuning (SFT) and model merging. We integrated reasoning chains into the SFT dataset, enabling AstroSage-Llama-3.1-70B to either answer the user query immediately, or first emit a human-readable thought process. Evaluated on a validated subset of 3,846 questions from the AstroMLab-1 benchmark (Ting et al., 2024) -- derived from literature withheld during training -- AstroSage-Llama-3.1-70B achieves top-tier performance (89.0%), matching GPT-5.2, Claude-4.5-Opus, and Gemini-3-Pro while being more cost-efficient. This work demonstrates that domain specialization, when applied to large-scale models, can enable them to outperform generalist counterparts in specialized knowledge areas like astronomy, thereby advancing the frontier of AI capabilities in the field. △ Less

Submitted 19 February, 2026; v1 submitted 23 May, 2025; originally announced May 2025.

AUTOLAW: Enhancing Legal Compliance in Large Language Models via Case Law Generation and Jury-Inspired Deliberation

Authors: Tai D. Nguyen, Long H. Pham, Jun Sun

Abstract: The rapid advancement of domain-specific large language models (LLMs) in fields like law necessitates frameworks that account for nuanced regional legal distinctions, which are critical for ensuring compliance and trustworthiness. Existing legal evaluation benchmarks often lack adaptability and fail to address diverse local contexts, limiting their utility in dynamically evolving regulatory landsc… ▽ More The rapid advancement of domain-specific large language models (LLMs) in fields like law necessitates frameworks that account for nuanced regional legal distinctions, which are critical for ensuring compliance and trustworthiness. Existing legal evaluation benchmarks often lack adaptability and fail to address diverse local contexts, limiting their utility in dynamically evolving regulatory landscapes. To address these gaps, we propose AutoLaw, a novel violation detection framework that combines adversarial data generation with a jury-inspired deliberation process to enhance legal compliance of LLMs. Unlike static approaches, AutoLaw dynamically synthesizes case law to reflect local regulations and employs a pool of LLM-based "jurors" to simulate judicial decision-making. Jurors are ranked and selected based on synthesized legal expertise, enabling a deliberation process that minimizes bias and improves detection accuracy. Evaluations across three benchmarks: Law-SG, Case-SG (legality), and Unfair-TOS (policy), demonstrate AutoLaw's effectiveness: adversarial data generation improves LLM discrimination, while the jury-based voting strategy significantly boosts violation detection rates. Our results highlight the framework's ability to adaptively probe legal misalignments and deliver reliable, context-aware judgments, offering a scalable solution for evaluating and enhancing LLMs in legally sensitive applications. △ Less

Submitted 19 June, 2025; v1 submitted 20 May, 2025; originally announced May 2025.

Quantized Approximate Signal Processing (QASP): Towards Homomorphic Encryption for audio

Authors: Tu Duyen Nguyen, Adrien Lesage, Clotilde Cantini, Rachid Riad

Abstract: Audio and speech data are increasingly used in machine learning applications such as speech recognition, speaker identification, and mental health monitoring. However, the passive collection of this data by audio listening devices raises significant privacy concerns. Fully homomorphic encryption (FHE) offers a promising solution by enabling computations on encrypted data and preserving user privac… ▽ More Audio and speech data are increasingly used in machine learning applications such as speech recognition, speaker identification, and mental health monitoring. However, the passive collection of this data by audio listening devices raises significant privacy concerns. Fully homomorphic encryption (FHE) offers a promising solution by enabling computations on encrypted data and preserving user privacy. Despite its potential, prior attempts to apply FHE to audio processing have faced challenges, particularly in securely computing time frequency representations, a critical step in many audio tasks. Here, we addressed this gap by introducing a fully secure pipeline that computes, with FHE and quantized neural network operations, four fundamental time-frequency representations: Short-Time Fourier Transform (STFT), Mel filterbanks, Mel-frequency cepstral coefficients (MFCCs), and gammatone filters. Our methods also support the private computation of audio descriptors and convolutional neural network (CNN) classifiers. Besides, we proposed approximate STFT algorithms that lighten computation and bit use for statistical and machine learning analyses. We ran experiments on the VocalSet and OxVoc datasets demonstrating the fully private computation of our approach. We showed significant performance improvements with STFT approximation in private statistical analysis of audio markers, and for vocal exercise classification with CNNs. Our results reveal that our approximations substantially reduce error rates compared to conventional STFT implementations in FHE. We also demonstrated a fully private classification based on the raw audio for gender and vocal exercise classification. Finally, we provided a practical heuristic for parameter selection, making quantized approximate signal processing accessible to researchers and practitioners aiming to protect sensitive audio data. △ Less

Submitted 15 May, 2025; originally announced May 2025.

Comments: 34 pages, 5 figures

A large-scale evaluation of commonsense knowledge in humans and large language models

Authors: Tuan Dung Nguyen, Duncan J. Watts, Mark E. Whiting

Abstract: Commonsense knowledge, a major constituent of artificial intelligence (AI), is primarily evaluated in practice by human-prescribed ground-truth labels. An important, albeit implicit, assumption of these labels is that they accurately capture what any human would think, effectively treating human common sense as homogeneous. However, recent empirical work has shown that humans vary enormously in wh… ▽ More Commonsense knowledge, a major constituent of artificial intelligence (AI), is primarily evaluated in practice by human-prescribed ground-truth labels. An important, albeit implicit, assumption of these labels is that they accurately capture what any human would think, effectively treating human common sense as homogeneous. However, recent empirical work has shown that humans vary enormously in what they consider commonsensical; thus what appears self-evident to one benchmark designer may not be so to another. Here, we propose a method for assessing commonsense knowledge in AI, specifically in large language models (LLMs), that incorporates empirically observed heterogeneity among humans by measuring the correspondence between a model's judgment and that of a human population. We first find that, when treated as independent survey respondents, most LLMs remain below the human median in their individual commonsense competence. Second, when used as simulators of a hypothetical population, LLMs correlate with real humans only modestly in the extent to which they agree on the same set of statements. In both cases, smaller, open-weight models are surprisingly more competitive than larger, proprietary frontier models. Our evaluation framework, which ties commonsense knowledge to its cultural basis, contributes to the growing call for adapting AI models to human collectivities that possess different, often incompatible, social stocks of knowledge. △ Less

Submitted 21 January, 2026; v1 submitted 15 May, 2025; originally announced May 2025.

Comments: Code and data: https://github.com/Watts-Lab/commonsense-llm-eval

Neural Brain: A Neuroscience-inspired Framework for Embodied Agents

Authors: Jian Liu, Xiongtao Shi, Thai Duy Nguyen, Haitian Zhang, Tianxiang Zhang, Wei Sun, Yanjie Li, Athanasios V. Vasilakos, Giovanni Iacca, Arshad Ali Khan, Arvind Kumar, Jae Won Cho, Ajmal Mian, Lihua Xie, Erik Cambria, Lin Wang

Abstract: The rapid evolution of artificial intelligence (AI) has shifted from static, data-driven models to dynamic systems capable of perceiving and interacting with real-world environments. Despite advancements in pattern recognition and symbolic reasoning, current AI systems, such as large language models, remain disembodied, unable to physically engage with the world. This limitation has driven the ris… ▽ More The rapid evolution of artificial intelligence (AI) has shifted from static, data-driven models to dynamic systems capable of perceiving and interacting with real-world environments. Despite advancements in pattern recognition and symbolic reasoning, current AI systems, such as large language models, remain disembodied, unable to physically engage with the world. This limitation has driven the rise of embodied AI, where autonomous agents, such as humanoid robots, must navigate and manipulate unstructured environments with human-like adaptability. At the core of this challenge lies the concept of Neural Brain, a central intelligence system designed to drive embodied agents with human-like adaptability. A Neural Brain must seamlessly integrate multimodal sensing and perception with cognitive capabilities. Achieving this also requires an adaptive memory system and energy-efficient hardware-software co-design, enabling real-time action in dynamic environments. This paper introduces a unified framework for the Neural Brain of embodied agents, addressing two fundamental challenges: (1) defining the core components of Neural Brain and (2) bridging the gap between static AI models and the dynamic adaptability required for real-world deployment. To this end, we propose a biologically inspired architecture that integrates multimodal active sensing, perception-cognition-action function, neuroplasticity-based memory storage and updating, and neuromorphic hardware/software optimization. Furthermore, we also review the latest research on embodied agents across these four aspects and analyze the gap between current AI systems and human intelligence. By synthesizing insights from neuroscience, we outline a roadmap towards the development of generalizable, autonomous agents capable of human-level intelligence in real-world scenarios. △ Less

Submitted 6 October, 2025; v1 submitted 12 May, 2025; originally announced May 2025.

Comments: 51 pages, 17 figures, 9 tables

The Mean of Multi-Object Trajectories

Authors: Tran Thien Dat Nguyen, Ba Tuong Vo, Ba-Ngu Vo, Hoa Van Nguyen, Changbeom Shim

Abstract: This paper introduces the concept of a mean for trajectories and multi-object trajectories (defined as sets or multi-sets of trajectories) along with algorithms for computing them. Specifically, we use the Fréchet mean, and metrics based on the optimal sub-pattern assignment (OSPA) construct, to extend the notion of average from vectors to trajectories and multi-object trajectories. Further, we de… ▽ More This paper introduces the concept of a mean for trajectories and multi-object trajectories (defined as sets or multi-sets of trajectories) along with algorithms for computing them. Specifically, we use the Fréchet mean, and metrics based on the optimal sub-pattern assignment (OSPA) construct, to extend the notion of average from vectors to trajectories and multi-object trajectories. Further, we develop efficient algorithms to compute these means using greedy search and Gibbs sampling. Using distributed multi-object tracking as an application, we demonstrate that the Fréchet mean approach to multi-object trajectory consensus significantly outperforms state-of-the-art distributed multi-object tracking methods. △ Less

Submitted 22 February, 2026; v1 submitted 28 April, 2025; originally announced April 2025.

Journal ref: in IEEE Transactions on Signal Processing, vol. 74, pp. 531-544, 2026

Autonomous and Self-Adapting System for Synthetic Media Detection and Attribution

Authors: Aref Azizpour, Tai D. Nguyen, Matthew C. Stamm

Abstract: Rapid advances in generative AI have enabled the creation of highly realistic synthetic images, which, while beneficial in many domains, also pose serious risks in terms of disinformation, fraud, and other malicious applications. Current synthetic image identification systems are typically static, relying on feature representations learned from known generators; as new generative models emerge, th… ▽ More Rapid advances in generative AI have enabled the creation of highly realistic synthetic images, which, while beneficial in many domains, also pose serious risks in terms of disinformation, fraud, and other malicious applications. Current synthetic image identification systems are typically static, relying on feature representations learned from known generators; as new generative models emerge, these systems suffer from severe performance degradation. In this paper, we introduce the concept of an autonomous self-adaptive synthetic media identification system -- one that not only detects synthetic images and attributes them to known sources but also autonomously identifies and incorporates novel generators without human intervention. Our approach leverages an open-set identification strategy with an evolvable embedding space that distinguishes between known and unknown sources. By employing an unsupervised clustering method to aggregate unknown samples into high-confidence clusters and continuously refining its decision boundaries, our system maintains robust detection and attribution performance even as the generative landscape evolves. Extensive experiments demonstrate that our method significantly outperforms existing approaches, marking a crucial step toward universal, adaptable forensic systems in the era of rapidly advancing generative models. △ Less

Submitted 19 June, 2026; v1 submitted 4 April, 2025; originally announced April 2025.

Forensic Self-Descriptions Are All You Need for Zero-Shot Detection, Open-Set Source Attribution, and Clustering of AI-generated Images

Authors: Tai D. Nguyen, Aref Azizpour, Matthew C. Stamm

Abstract: The emergence of advanced AI-based tools to generate realistic images poses significant challenges for forensic detection and source attribution, especially as new generative techniques appear rapidly. Traditional methods often fail to generalize to unseen generators due to reliance on features specific to known sources during training. To address this problem, we propose a novel approach that exp… ▽ More The emergence of advanced AI-based tools to generate realistic images poses significant challenges for forensic detection and source attribution, especially as new generative techniques appear rapidly. Traditional methods often fail to generalize to unseen generators due to reliance on features specific to known sources during training. To address this problem, we propose a novel approach that explicitly models forensic microstructures - subtle, pixel-level patterns unique to the image creation process. Using only real images in a self-supervised manner, we learn a set of diverse predictive filters to extract residuals that capture different aspects of these microstructures. By jointly modeling these residuals across multiple scales, we obtain a compact model whose parameters constitute a unique forensic self-description for each image. This self-description enables us to perform zero-shot detection of synthetic images, open-set source attribution of images, and clustering based on source without prior knowledge. Extensive experiments demonstrate that our method achieves superior accuracy and adaptability compared to competing techniques, advancing the state of the art in synthetic media forensics. △ Less

Submitted 26 March, 2025; originally announced March 2025.

Comments: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2025

MVFNet: Multipurpose Video Forensics Network using Multiple Forms of Forensic Evidence

Authors: Tai D. Nguyen, Matthew C. Stamm

Abstract: While videos can be falsified in many different ways, most existing forensic networks are specialized to detect only a single manipulation type (e.g. deepfake, inpainting). This poses a significant issue as the manipulation used to falsify a video is not known a priori. To address this problem, we propose MVFNet - a multipurpose video forensics network capable of detecting multiple types of manipu… ▽ More While videos can be falsified in many different ways, most existing forensic networks are specialized to detect only a single manipulation type (e.g. deepfake, inpainting). This poses a significant issue as the manipulation used to falsify a video is not known a priori. To address this problem, we propose MVFNet - a multipurpose video forensics network capable of detecting multiple types of manipulations including inpainting, deepfakes, splicing, and editing. Our network does this by extracting and jointly analyzing a broad set of forensic feature modalities that capture both spatial and temporal anomalies in falsified videos. To reliably detect and localize fake content of all shapes and sizes, our network employs a novel Multi-Scale Hierarchical Transformer module to identify forensic inconsistencies across multiple spatial scales. Experimental results show that our network obtains state-of-the-art performance in general scenarios where multiple different manipulations are possible, and rivals specialized detectors in targeted scenarios. △ Less

Submitted 26 March, 2025; originally announced March 2025.

Comments: Proceedings of the Winter Conference on Applications of Computer Vision (WACV) 2025

Advances in Protein Representation Learning: Methods, Applications, and Future Directions

Authors: Viet Thanh Duy Nguyen, Truong-Son Hy

Abstract: Proteins are complex biomolecules that play a central role in various biological processes, making them critical targets for breakthroughs in molecular biology, medical research, and drug discovery. Deciphering their intricate, hierarchical structures, and diverse functions is essential for advancing our understanding of life at the molecular level. Protein Representation Learning (PRL) has emerge… ▽ More Proteins are complex biomolecules that play a central role in various biological processes, making them critical targets for breakthroughs in molecular biology, medical research, and drug discovery. Deciphering their intricate, hierarchical structures, and diverse functions is essential for advancing our understanding of life at the molecular level. Protein Representation Learning (PRL) has emerged as a transformative approach, enabling the extraction of meaningful computational representations from protein data to address these challenges. In this paper, we provide a comprehensive review of PRL research, categorizing methodologies into five key areas: feature-based, sequence-based, structure-based, multimodal, and complex-based approaches. To support researchers in this rapidly evolving field, we introduce widely used databases for protein sequences, structures, and functions, which serve as essential resources for model development and evaluation. We also explore the diverse applications of these approaches in multiple domains, demonstrating their broad impact. Finally, we discuss pressing technical challenges and outline future directions to advance PRL, offering insights to inspire continued innovation in this foundational field. △ Less

Submitted 8 May, 2025; v1 submitted 20 March, 2025; originally announced March 2025.

The structure of polynomial growth for tree automata/transducers and MSO set queries

Authors: Paul Gallot, Nathan Lhote, Lê Thành Dũng Nguyên

Abstract: Given an $\mathbb{N}$-weighted tree automaton, we give a decision procedure for exponential vs polynomial growth (with respect to the input size) in quadratic time, and an algorithm that computes the exact polynomial degree of growth in cubic time. As a special case, they apply to the growth of the ambiguity of a nondeterministic tree automaton, i.e. the number of distinct accepting runs over a gi… ▽ More Given an $\mathbb{N}$-weighted tree automaton, we give a decision procedure for exponential vs polynomial growth (with respect to the input size) in quadratic time, and an algorithm that computes the exact polynomial degree of growth in cubic time. As a special case, they apply to the growth of the ambiguity of a nondeterministic tree automaton, i.e. the number of distinct accepting runs over a given input. We deduce analogous decidability results (ignoring complexity) for the growth of the number of results of set queries in Monadic Second-Order logic (MSO) over ranked trees. In the case of polynomial growth of degree $k$, we also prove a reparameterization theorem for such queries: their results can be mapped to $k$-tuples of input nodes in a finite-to-one and MSO-definable fashion. We then apply these tools to study growth rates and subclass membership problems for tree-to-tree functions. Using new proof strategies, we recover and generalize known results concerning polyregular functions, total deterministic macro tree transducers, and partial nondeterministic top-down tree transducers. In particular, we give a procedure to decide polynomial size-to-height increase for both macro tree transducers and MSO set interpretations, and compute the degree. The paper concludes with a survey of a wide range of related work. △ Less

Submitted 6 January, 2026; v1 submitted 17 January, 2025; originally announced January 2025.

Comments: 50 pages, major revision after phase 2 of TheoretiCS reviews. New in v4: results and references on nondeterministic top-down tree transducers, more improvements to size-to-height increase, appendix with more proof details on Sections 2 and 3, various typos and polishing