過去のお知らせ
2026年10月23日～29日に神戸で開催された2026 IEEE 15th Global Conference on Consumer Electronics (IEEE GCCE2026)に、当研究室より下記の14件の発表を行いました
また、修士1年生の小野塚君がExcellent Poster Award, Silver Prizeを、学部4年生の前田さんがExcellent Poster Award,Outstanding Prizeを受賞しました。
(1) Does Side Really Matter? Investigating Left-Right Bias in Skeleton-Based Human Action Recognition
Tatsuki Yamada, Ryota Goka, Keisuke Maeda, Naoki Saito, Takahiro Ogawa and Miki Haseyama
(2) Local Temporal Onset Density Control in Symbolic Music Arrangement via Sparse Autoencoders
Genki Masuyama, Keigo Sakurai, Ren Togo, Takahiro Ogawa and Miki Haseyama
(3) Quantifying Task Difficulty for LLMs Based on Reasoning Trajectories
Takuya Kinoshita, Ren Togo, Keisuke Maeda, Takahiro Ogawa and Miki Haseyama
(4) A Privacy-Preserving Unlearning Framework Based on Pixel-Level Encryption
Ibuki Kameya, Tasuku Nakajima, Keisuke Maeda, Takahiro Ogawa and Miki Haseyama
(5) Mitigating Concept Bleed-Through in Continual Personalized Image Generation via Latent Space-Based Target Selection
Toma Onozuka, Koshiro Toishi, Keisuke Maeda, Takahiro Ogawa and Miki Haseyama
(6) Spatial Audio Reproduction from 3D Gaussian Splatting Scenes via Mesh Conversion and Mesh-Based Acoustic Simulation
Akari Maeda, Masaki Yoshida, Ren Togo, Takahiro Ogawa and Miki Haseyama
(7) Efficient Linear-Probe Dataset Distillation for Large-Scale Pre-Trained Vision Models
Bincheng Peng, Guang Li, Ping Liu, Takahiro Ogawa and Miki Haseyama
(8) daptive Multi-Level Response Matching for Point Cloud Dataset Distillation
Ge Tian, Guang Li, Takahiro Ogawa and Miki Haseyama
(9) Impact of Expert-Following Strategies in Financial Asset Recommendation
Ryuki Unno, Koshi Watanabe, Keigo Sakurai, Keisuke Maeda, Takahiro Ogawa and Miki Haseyama
(10) Instruction-Tuned Vision Language Model for Bridge Inspection with Reason-Based Prompts
Seiya Kidoguchi, Keisuke Maeda, Takahiro Ogawa and Miki Haseyama
(11) owards Low-Power Time-Series Anomaly Detection with Spike-Based Variational Autoencoder
Akira Hasegawa, Kenta Kubota, Ren Togo, Takahiro Ogawa and Miki Haseyama
(12) Application of Deep Autoencoding Gaussian Mixture Model for Anomaly Detection in Hyperspectral Analysis of Rubber Materials
Kentaro Oishi, Ren Togo, Takahiro Ogawa and Miki Haseyama
(13) Ranking Refinement with Context-Dependent Acoustic Preference for Music Recommendation
Yuki Iwasa, Keigo Sakurai, Ren Togo, Takahiro Ogawa and Miki Haseyama
(14) Prompt Refinement for Vision-Language-Action Models Under Incomplete Instructions for Robot Control
Shin Kawamura, Tatsuki Seino, Naoki Saito, Ren Togo, Takahiro Ogawa and Miki Haseyama
