The University at Buffalo is a flagship campus of the State University of New York system (SUNY), excelling in engineering, AI, computing, biomedical sciences, medicine, and public health. It is a major center for research and innovation in upstate New York. It fosters research that advances regional and global impact. NECLA partnered with the University at Buffalo on research into adversarial training for visual content generation on large-scale vision systems, data-efficient learning, and multimodal AI for healthcare. We contributed to the refinement of dual projection GANs through collaboration, improving realism and diversity in synthesized images, with implications for both biometric security and creative applications. Please read about our latest news and collaborative publications with the State University of New York at Buffalo.
NEC Labs America is heading to Denver for CVPR 2026, one of the most prestigious gatherings in computer vision, machine learning, and pattern recognition. The IEEE/CVF Conference on Computer Vision and Pattern Recognition brings innovators from around the world to share breakthroughs.
Machine learning is shifting from learning from data alone to learning from both data and teacher models. Beta-KD uses uncertainty-aware Bayesian weighting to train compact multimodal AI without blindly trusting every teacher signal.
Real-world deployment requires sound event and acoustic scene classification systems to remain reliable in noisy, diverse environments on resource-constrained devices. Although contrastive language-audio pretraining (CLAP) models with Transformer-based audio encoders achieve strong zero-shot performance, their computational cost hinders deployment. In this paper, we propose Mix-CLAP, a computationally efficient, noise-aware CLAP model with knowledge-distilled audio encoders. Our method includes: (1) a two-stage knowledge distillation from teacher embeddings to two lightweight student encoders?one on clean audio, the other on noisy audio, and (2) adaptive inference that combines their embeddings together with a fusion parameter and minimizes the parameterized entropy at test time. Experiments show that Mix-CLAP with MobileNetV3-based audio encoders greatly improves computational efficiency, while achieving a comparable average accuracy of 52.58% to the Transformer-based CLAP model at 52.83% on the recorded ESC50 datasets with different devices including microphones and fiber-optic distributed acoustic sensors under diverse conditions, making it suitable for real-world, resource-constrained applications.
Knowledge distillation establishes a learning paradigm that leverages both data supervision and teacher guidance. However, determining the optimal balance between learning from data and learning from the teacher is challenging, as some samples may be noisy while others are subject to teacher uncertainty. This motivates the need for adaptively balancing data and teacher supervision. We propose Beta-weighted Knowledge Distillation (Beta-KD), an uncertainty-aware distillation framework that adaptively modulates how much the student relies on teacher guidance. Specifically, we formulate teacher–student learning from a unified Bayesian perspective and interpret teacher supervision as a Gibbs prior over student activations. This yields a closed-form, uncertainty-aware weighting mechanism and supports arbitrary distillation objectives and their combinations. Extensive experiments on multimodal VQA benchmarks demonstrate that distilling student Vision-Language Models from a large teacher VLM consistently improves performance. The results show that Beta-KD outperforms existing knowledge distillation methods.
On National Intern Day, NEC Laboratories America celebrates the bright minds shaping tomorrow’s technology. Each summer, interns from top universities work side-by-side with our researchers on real-world challenges in AI, cybersecurity, data science, and more. From groundbreaking research to team-building events, our interns contribute fresh ideas and bold thinking that power NEC’s innovation engine.
Contrastive Language-Audio Pretraining (CLAP) models have demonstrated unprecedented performance in various acoustic signal recognition tasks. Fiber-optic-based acoustic recognition is one of the most important downstream tasks and plays a significant role in environmental sensing. Adapting CLAP for fiber-optic acoustic recognition has become an active research area. As a non-conventional acoustic sensor, fiberoptic acoustic recognition presents a challenging, domain-specific, low-shot deployment environment with significant domain shifts due to unique frequency response and noise characteristics. To address these challenges, we propose a support-based adaptation method, CLAP-S, which linearly interpolates a CLAP Adapter with the Support Set, leveraging both implicit knowledge through fine-tuning and explicit knowledge retrieved from memory for cross-domain generalization. Experimental results show that our method delivers competitive performance on both laboratory recorded fiber-optic ESC-50 datasets and a real-world fiber optic gunshot-firework dataset. Our research also provides valuable insights for other downstream acoustic recognition tasks.
Contrastive Language-Audio Pretraining (CLAP) models have demonstrated unprecedented performance in various acoustic signal recognition tasks. Fiber optic-based acoustic recognition is one of the most important downstream tasks and plays a significant role in environmental sensing. Adapting CLAP for fiber-optic acoustic recognition has become an active research area. As a non-conventional acoustic sensor, fiber-optic acoustic recognition presents a challenging, domain-specific, low-shot deployment environment with significant domain shifts due to unique frequency response and noise characteristics. To address these challenges, we propose a support-based adaptation method, CLAP-S, which linearly interpolates a CLAP Adapter with the Support Set, leveraging both implicit knowledge through fine-tuning and explicit knowledge retrieved from memory for cross-domain generalization. Experimental results show that our method delivers competitive performance on both laboratory-recorded fiber-optic ESC-50 datasets and a real-world fiber-optic gunshot-firework dataset. Our research also provides valuable insights for other downstream acoustic recognition tasks.
Forget passwords—identity verification can now be accomplished with the touch of a finger or in the blink of an eye as the biometrics field expands to encompass new techniques and application areas.
Biometric authentication is the recognition of human identity via unique anatomical features. The development of novel methods parallels widespread application by consumer devices, law enforcement, and access control. In particular, methods based on finger veins, as compared to face and fingerprints, obviate privacy concerns and degradation due to wear, age, and obscuration. However, they are two-dimensional (2D) and are fundamentally limited by conventional imaging and tissue-light scattering. In this work, for the first time, to the best of our knowledge, we demonstrate a method of three-dimensional (3D) finger vein biometric authentication based on photoacoustic tomography. Using a compact photoacoustic tomography setup and a novel recognition algorithm, the advantages of 3D are demonstrated via biometric authentication of index finger vessels with false acceptance, false rejection, and equal error rates <1.23%, <9.27%, and <0.13%, respectively, when comparing one finger, a false acceptance rate improvement >10× when comparing multiple fingers, and <0.7% when rotating fingers ±30.
The advent of LTE into the unlicensed spectrum has necessitated the understanding of its operational efficiency when sharing spectrum with different radio access technologies. Our study reveals that LTE, owing to its inherent transmission characteristics, suffers significant performance degradation in the presence of interference caused by hidden terminals. This motivates the need for interference-awareness in LTE’s channel access in unlicensed spectrum. To address this problem, we propose ELI. ELI’s three-pronged solution equips the LTE base station with novel techniques to: (a) accurately detect and measure interference caused by hidden terminals, (b) collect interference statistics from clients across different channels with affordable overhead, and (c) leverage interference-awareness to improve its channel access performance. Our evaluations show that ELI can achieve 1.5-2x throughput gains over baseline schemes. Finally, ELI is LTE-LAA/MulteFire-standard compliant and can be deployed over the existing LTE-LAA implementation without any modifications.
Princeton Office
4 Independence Way, Suite 200
Princeton, NJ 08540
+1 609-520-1555
San Jose Office
2033 Gateway Place, Suite 200
San Jose, CA 95110
+1 408-863-6007
NEC Laboratories America, Inc. (NEC Labs) is the US-based center for NEC Corporation’s global network of corporate research laboratories. Our diverse research groups collaborate with industry, academia and governments to provide disruptive solutions to complex problems. A leader in the integration of IT and network technologies with more than 100 years of expertise, NEC provides a combination of products and solutions that cross-utilize the company’s experience and global resources to meet the complex and ever-changing needs of its customers.