Look Less, Think Faster: Joint Token-Compute Adaptation for Multimodal LLMs (2026)

Venue: ECCV, 2026.

What CLIP Knows but Cannot Say: Recovering Negation from Frozen Intermediate Features (2026)

Venue: ECCV ORAL, 2026 [Top 2.3%].

Learning to Inference Adaptively for Multimodal Large Language Models (2025)

Proposes an inference-time policy that picks per-query settings—image resolution, token budget, modality routing, early-exit—to trade accuracy, latency, and energy on the fly. Yields higher throughput at fixed utility and graceful quality under tight budgets across representative MLLMs.
Venue: IEEE/CVF International Conference on Computer Vision (ICCV), 2025.
Acceptance Rate: 24.0% (2,699 / 11,239)
Keywords: adaptive inference, MLLM, dynamic modality routing, early exit, token budgeting, latency-energy trade-offs.

SKALD: Learning-Based Shot Assembly for Coherent Multi-Shot Video Creation (2025)

Introduces a learned clip-coherence objective that scores spatial/temporal/semantic flow and a beam-search assembler to select coherent shot sequences under style/length constraints. Outperforms rule-based and caption-similarity baselines while remaining plug-and-play with common editors/generators.
Venue: IEEE/CVF International Conference on Computer Vision (ICCV), 2025.
Acceptance Rate: 24.0% (2,699 / 11,239)
Keywords: video assembly, coherence, shot selection, beam search, generative editing.

Improving Semi-Supervised Semantic Segmentation with Sliced-Wasserstein Feature Alignment and Uniformity (2025)

Aligns labeled/unlabeled features via Sliced-Wasserstein distance and adds a uniformity prior to avoid cluster collapse, boosting mIoU under extreme label sparsity and class imbalance. Compatible with modern backbones and stable across confidence thresholds.
Venue: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).
Acceptance Rate: 22.1% (2,878 / 13,008)
Keywords: semi-supervised segmentation, feature alignment, Sliced-Wasserstein, uniformity, class imbalance.

Agile3D: Adaptive Contention- and Content-Aware 3D Object Detection for Embedded GPUs (2025)

Right-sizes LiDAR detection by reacting jointly to scene content and on-device contention, adjusting resolution/blocks to hold tail latency and energy in check with minimal AP loss. Demonstrates portable gains across embedded platforms with a lightweight runtime.
Venue: ACM International Conference on Mobile Systems, Applications, and Services (MobiSys).
Acceptance Rate: 18.0% (42 / 233)
Keywords: 3D detection, embedded GPU, adaptive scheduling, contention-aware, LiDAR.

ReCON: Training-Free Acceleration for Text-to-Image Synthesis with Retrieval of Concept Prompt Trajectories (2024)

Speeds diffusion generation by reusing retrieved concept trajectories—preserving fidelity and style without retraining—and complements existing schedulers. Works across diverse prompts and models with consistent step reductions.
Venue: European Conference on Computer Vision (ECCV).
Acceptance Rate: 27.9% (2,395 / 8,585)
Keywords: diffusion, retrieval-augmented generation, prompt trajectories, speed-ups.

SENSORBFT: Energy-Efficient Fault-Tolerant Target Localization (2024)

Uses lightweight consensus with sensing to localize under Byzantine faults, lowering message cost and energy while maintaining accuracy. Scales to dense deployments with bounded overhead.
Venue: IEEE International Conference on Distributed Computing Systems (ICDCS).
Acceptance Rate: 21.9% (121 accepted; ≈553 submissions)
Keywords: federated sensing, Byzantine resilience, consensus, energy efficiency.

Adaptive Object Detection algorithms for Resource Constrained Autonomous Robotic Systems (2024)

Extends Virtuoso for real-time detection under contention by parallelizing detection and tracking and using multi-resolution tracker initialization matched to box size—maintains throughput even with costlier detectors. Validated on embedded hardware.
Venue: SPIE Defense + Commercial Sensing, Proc. SPIE Vol. 13058, Disruptive Technologies in Information Sciences VIII (SPIE DCS) — Collaboration with the Army Research Lab stemming from internship work; includes a working software artifact.
Keywords: robotics, multi-threaded vision, tracker initialization, real-time.

Asgard: Are NoSQL databases suitable for ephemeral data in serverless workloads? (2023)

Empirical study of NoSQL stores for short-lived, bursty serverless data paths; analyzes throughput-latency-cost under TTL/ephemeral patterns. Provides configuration guidance for scaling transient state without overprovisioning.
Venue: Frontiers in High Performance Computing (Frontiers HPC) — Joint work with an undergraduate and an alum now at Salesforce; perspective-style article with early results on “serverless cloud computing meets NoSQL” for volatile workloads.
Keywords: serverless, ephemeral data, NoSQL, cost-performance, TTL.

How to Learn Collaboratively - Federated Learning to Peer-to-Peer Learning and What's At Stake (2023)

Maps FL ↔ P2P trade-offs (privacy, personalization, robustness, communication cost) and proposes deployment metrics and design patterns. Highlights hybrid topologies (hub-and-spoke + peer gossip) and reliability/auditing gaps.
Venue: IEEE/IFIP International Conference on Dependable Systems and Networks — DISrupt Workshop (DSN-DISrupt) — “Disruptive paper” outlining our hub-and-spoke decentralized learning approach: higher availability and privacy than federated learning (FL) and lower compute/communication overhead than peer-to-peer (P2P) learning—best of both worlds.
Keywords: federated learning, peer-to-peer, collaboration, privacy, robustness.

Vega: Drone-based Multi-Altitude Target Detection for Autonomous Surveillance (2023)

Fuses complementary views from different flight altitudes and schedules sensing to balance coverage, energy, and detection confidence. Improves robustness to occlusions/scale changes under onboard compute limits.
Venue: IEEE International Conference on Distributed Computing in Sensor Systems and Internet of Things (DCOSS-IoT).
Acceptance Rate: 23% (20 / 86)
Keywords: UAV sensing, multi-altitude fusion, energy-aware scheduling, surveillance.

FLAIR: Defense against Model Poisoning Attack in Federated Learning (2023)

Combines anomaly scoring with robust aggregation to down-weight poisoned updates while preserving benign personalization. Sustains accuracy across heterogeneous clients and attack strengths with modest overhead.
Venue: ACM Asia Conference on Computer and Communications Security (AsiaCCS).
Acceptance Rate: 18% (418 / 2,322)
Keywords: federated learning, poisoning defense, robust aggregation, anomaly detection.

Virtuoso: Energy- and Latency-Aware Streamlining of Streaming Videos on SOCs (2022)

Enforces energy SLOs by jointly scheduling operators, resolutions, and model variants to keep latency within budget on mobile SoCs; includes deployable runtime and open implementation. Evaluations show stable QoE under tight power caps.
Venue: ACM Transactions on Design Automation of Electronic Systems (ACM TODAES).
Keywords: mobile SoC, energy SLO, operator scheduling, video pipelines.

SMARTADAPT: Multi-branch Object Detection Framework for Videos on Mobiles (2022)

Switches among detection branches based on content dynamics, improving accuracy-latency trade-offs for on-device video analytics. Shows gains across varied motion/texture regimes.
Venue: IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).
Acceptance Rate: 25.3% (2,064 / 8,161)
Keywords: adaptive detection, mobile vision, content-aware routing, on-device.