6 papers
A 1000-hour EEG-EMG-audio dataset of Japanese speech production
Motoshige Sato, Ilya Horiguchi, Masakazu Inoue +6
We present a multimodal dataset of 1020 hours of simultaneously recorded scalp electroencephalography (EEG), facial electromyography (EMG), and speech audio from three healthy nati…
MCM: Multi-layer Concept Map for Efficient Concept Learning from Masked Images
Yuwei Sun, Lu Mi, Ippei Fujisawa +4
Masking strategies commonly employed in natural language processing are still underexplored in vision tasks such as concept learning, where conventional methods typically rely on f…
Measuring How LLMs Internalize Human Psychological Concepts: A preliminary analysis
Hiro Taiyo Hamada, Ippei Fujisawa, Genji Kawakita +1
Large Language Models (LLMs) such as ChatGPT have shown remarkable abilities in producing human-like text. However, it is unclear how accurately these models internalize concepts t…
Decoding Vision Transformers: the Diffusion Steering Lens
Ryota Takatsuki, Sonia Joseph, Ippei Fujisawa +1
Logit Lens is a widely adopted method for mechanistic interpretability of transformer-based language models, enabling the analysis of how internal representations evolve across lay…
ProcBench: Benchmark for Multi-Step Reasoning and Following Procedure
Ippei Fujisawa, Sensho Nobe, Hiroki Seto +5
Reasoning is central to a wide range of intellectual activities, and while the capabilities of large language models (LLMs) continue to advance, their performance in reasoning task…
Remembering Transformer for Continual Learning
Yuwei Sun, Ippei Fujisawa, Arthur Juliani +2
Neural networks encounter the challenge of Catastrophic Forgetting (CF) in continual learning, where new task learning interferes with previously learned knowledge. Existing data f…