Selected work

Representative Papers

Explainability Boosted Anomaly Detection Framework for O-RAN based NextG Networks

Aug 14, 2026

This study addresses security vulnerabilities and anomaly detection challenges in next-generation cellular networks by proposing an explainable AI-based anomaly detection framework for O-RAN. By integrating post-hoc explanation methods to identify Key Performance Metrics (KPMs), the approach enables precise malicious traffic identification and attack characterization within a real-world O-RAN testbed. The proposed method reduces data complexity by 80% while maintaining high detection accuracy, effectively balancing computational efficiency, detection performance, and model interpretability. Consequently, this work significantly enhances the security resilience and practical deployment feasibility of NextG networks, offering a robust solution for identifying sophisticated threats in open radio access network environments without compromising operational efficiency or transparency.

0 citationsRead paper

Phase Marginalization for Patch-Grid Instability in Vision Transformers

Jun 06, 2026

Vision Transformers suffer from phase dependency in dense prediction tasks due to their fixed image patch grid, causing pixel-level outputs to vary unstably near patch boundaries. This work formally characterizes patch phase as a measurable nuisance variable and introduces Uniform Phase Marginalization—a plug-and-play, test-time post-processing method that requires no retraining. By performing forward inference across multiple structured phase shifts, inversely aligning the predictions, and aggregating them in the original image coordinate system, the method substantially improves prediction consistency. On Cityscapes, it surpasses the strongest four-shift test-time augmentation by 0.31 mIoU at comparable computational cost; further increasing the number of phases (K=8/16) yields only marginal gains while significantly increasing latency.

0 citationsRead paper

When Does Adaptive Guidance Help? Belief-Aware Privileged Distillation for Autonomous Driving Under Partial Observability

May 24, 2026

This work addresses the limitations of existing knowledge distillation methods in partially observable environments, where fixed distillation coefficients fail to account for agent uncertainty. The authors propose Belief-Aware Guided Soft Actor-Critic (BA-GSAC), which dynamically adjusts the distillation coefficient based on ensemble model disagreement and systematically evaluates multiple scheduling strategies under varying degrees of observation occlusion. Experiments demonstrate that adaptive guidance significantly improves training stability in mildly to moderately partial observability (coefficient of variation: 13.3% vs. 29.8%), though it tends to degrade to its minimum value under severe occlusion. In such high-occlusion regimes, a linear decay strategy achieves superior performance (mean return: 116.5, coefficient of variation: 8.9%). The study further uncovers an “observability blind spot” in ensemble-based approaches under heavy occlusion and suggests architectural improvements targeting full-state information as a promising direction.

0 citationsRead paper

LogRouter: Adaptive Two-Level LLM Routing for Log Question Answering in Big Data Systems

May 18, 2026

This work addresses the challenge of enabling low-cost natural language question answering over massive log streams in resource-constrained, self-hosted environments. The authors propose an end-to-end system featuring a training-free, two-tier cost-aware routing mechanism that dynamically selects one of four execution paths based on a keyword vocabulary. Within the semantic path, the system adaptively invokes either a 14B or 32B large language model and leverages a specialized Coder model to generate SQL queries. The architecture integrates PySpark with Drain3 for log parsing, GPU-accelerated embeddings, and dual-index storage using Apache Druid and PostgreSQL/pgvector. Evaluated on four LogHub datasets, the system achieves an average accuracy of 88.4% (reaching 94.7% on Linux logs), with an end-to-end latency of 18.6 seconds—55% lower than a fixed 32B baseline—and demonstrates significantly higher RAGAS Faithfulness compared to a fixed 14B baseline.

0 citationsRead paper

Performance Evaluation of Dual RIS-Assisted Received Space Shift Keying Modulation

Nov 23, 2025

To address the limited signal routing flexibility in single-RIS indoor systems caused by static reflection, this paper proposes a cooperative dual-reconfigurable intelligent surface (RIS) architecture. The first RIS provides baseline channel enhancement, while the second RIS (RIS₂) dynamically adjusts its reflection phases according to source data bits—enabling bit-driven physical-layer spatial modulation and beam-level signal routing. This work pioneers the integration of spatial shift keying (SSK) with dynamic phase mapping across two RISs, establishing an end-to-end transmission framework under a multi-hop channel model. Experimental results demonstrate significant improvements in achievable capacity and substantial reduction in outage probability across varying carrier frequencies and inter-RIS distances, thereby enabling high-accuracy, data-dependent intelligent indoor wireless coverage.

0 citationsRead paper

Recent publications

Latest Papers

Explainability Boosted Anomaly Detection Framework for O-RAN based NextG Networks

Aug 14, 2026

This study addresses security vulnerabilities and anomaly detection challenges in next-generation cellular networks by proposing an explainable AI-based anomaly detection framework for O-RAN. By integrating post-hoc explanation methods to identify Key Performance Metrics (KPMs), the approach enables precise malicious traffic identification and attack characterization within a real-world O-RAN testbed. The proposed method reduces data complexity by 80% while maintaining high detection accuracy, effectively balancing computational efficiency, detection performance, and model interpretability. Consequently, this work significantly enhances the security resilience and practical deployment feasibility of NextG networks, offering a robust solution for identifying sophisticated threats in open radio access network environments without compromising operational efficiency or transparency.

0 citationsRead paper

Phase Marginalization for Patch-Grid Instability in Vision Transformers

Jun 06, 2026

Vision Transformers suffer from phase dependency in dense prediction tasks due to their fixed image patch grid, causing pixel-level outputs to vary unstably near patch boundaries. This work formally characterizes patch phase as a measurable nuisance variable and introduces Uniform Phase Marginalization—a plug-and-play, test-time post-processing method that requires no retraining. By performing forward inference across multiple structured phase shifts, inversely aligning the predictions, and aggregating them in the original image coordinate system, the method substantially improves prediction consistency. On Cityscapes, it surpasses the strongest four-shift test-time augmentation by 0.31 mIoU at comparable computational cost; further increasing the number of phases (K=8/16) yields only marginal gains while significantly increasing latency.

0 citationsRead paper

When Does Adaptive Guidance Help? Belief-Aware Privileged Distillation for Autonomous Driving Under Partial Observability

May 24, 2026

This work addresses the limitations of existing knowledge distillation methods in partially observable environments, where fixed distillation coefficients fail to account for agent uncertainty. The authors propose Belief-Aware Guided Soft Actor-Critic (BA-GSAC), which dynamically adjusts the distillation coefficient based on ensemble model disagreement and systematically evaluates multiple scheduling strategies under varying degrees of observation occlusion. Experiments demonstrate that adaptive guidance significantly improves training stability in mildly to moderately partial observability (coefficient of variation: 13.3% vs. 29.8%), though it tends to degrade to its minimum value under severe occlusion. In such high-occlusion regimes, a linear decay strategy achieves superior performance (mean return: 116.5, coefficient of variation: 8.9%). The study further uncovers an “observability blind spot” in ensemble-based approaches under heavy occlusion and suggests architectural improvements targeting full-state information as a promising direction.

0 citationsRead paper

LogRouter: Adaptive Two-Level LLM Routing for Log Question Answering in Big Data Systems

May 18, 2026

This work addresses the challenge of enabling low-cost natural language question answering over massive log streams in resource-constrained, self-hosted environments. The authors propose an end-to-end system featuring a training-free, two-tier cost-aware routing mechanism that dynamically selects one of four execution paths based on a keyword vocabulary. Within the semantic path, the system adaptively invokes either a 14B or 32B large language model and leverages a specialized Coder model to generate SQL queries. The architecture integrates PySpark with Drain3 for log parsing, GPU-accelerated embeddings, and dual-index storage using Apache Druid and PostgreSQL/pgvector. Evaluated on four LogHub datasets, the system achieves an average accuracy of 88.4% (reaching 94.7% on Linux logs), with an end-to-end latency of 18.6 seconds—55% lower than a fixed 32B baseline—and demonstrates significantly higher RAGAS Faithfulness compared to a fixed 14B baseline.

0 citationsRead paper

Performance Evaluation of Dual RIS-Assisted Received Space Shift Keying Modulation

Nov 23, 2025

To address the limited signal routing flexibility in single-RIS indoor systems caused by static reflection, this paper proposes a cooperative dual-reconfigurable intelligent surface (RIS) architecture. The first RIS provides baseline channel enhancement, while the second RIS (RIS₂) dynamically adjusts its reflection phases according to source data bits—enabling bit-driven physical-layer spatial modulation and beam-level signal routing. This work pioneers the integration of spatial shift keying (SSK) with dynamic phase mapping across two RISs, establishing an end-to-end transmission framework under a multi-hop channel model. Experimental results demonstrate significant improvements in achievable capacity and substantial reduction in outage probability across varying carrier frequencies and inter-RIS distances, thereby enabling high-accuracy, data-dependent intelligent indoor wireless coverage.

0 citationsRead paper