10 papers
Zero-shot Concept Bottleneck Models
Shin'ya Yamaguchi, Kosuke Nishida, Daiki Chijiwa +1
Concept bottleneck models (CBMs) are inherently interpretable and intervenable neural network models, which explain their final label prediction by the intermediate prediction of h…
Parallel In-context Learning for Large Vision Language Models
Shin'ya Yamaguchi, Daiki Chijiwa, Tamao Sakao +1
Large vision-language models (LVLMs) employ multi-modal in-context learning (MM-ICL) to adapt to new tasks by leveraging demonstration examples. While increasing the number of demo…
Rationale-Enhanced Decoding for Multi-modal Chain-of-Thought
Shin'ya Yamaguchi, Kosuke Nishida, Daiki Chijiwa
Large vision-language models (LVLMs) have demonstrated remarkable capabilities by integrating pre-trained vision encoders with large language models (LLMs). Similar to single-modal…
Do We Really Need Permutations? Impact of Model Width on Linear Mode Connectivity
Akira Ito, Masanori Yamada, Daiki Chijiwa +1
Recently, Ainsworth et al. empirically demonstrated that, given two independently trained models, applying a parameter permutation that preserves the input-output behavior allows t…
Lossless Vocabulary Reduction for Auto-Regressive Language Models
Daiki Chijiwa, Taku Hasegawa, Kyosuke Nishida +4
Tokenization -- the process of decomposing a given text into a sequence of subwords called tokens -- is one of the key components in the development of language models. Particularl…
The Strong Lottery Ticket Hypothesis for Multi-Head Attention Mechanisms
Hikari Otsuka, Daiki Chijiwa, Yasuyuki Okoshi +3
The strong lottery ticket hypothesis (SLTH) conjectures that high-performing subnetworks, called strong lottery tickets (SLTs), are hidden in randomly initialized neural networks.…