2 citations · 5 across the 14 of their papers we have counts for
14 papers
Distilling Many-Shot In-Context Learning into a Cheat Sheet
Ukyo Honda, Soichiro Murakami, Peinan Zhang
Recent advances in large language models (LLMs) enable effective in-context learning (ICL) with many-shot examples, but at the cost of high computational demand due to longer input…
Exploring Explanations Improves the Robustness of In-Context Learning
Ukyo Honda, Tatsushi Oka
In-context learning (ICL) has emerged as a successful paradigm for leveraging large language models (LLMs). However, it often struggles to generalize beyond the distribution of the…
Exploring the Relationship Between Diversity and Quality in Ad Text Generation
Yoichi Aoki, Soichiro Murakami, Ukyo Honda +1
In natural language generation for advertising, creating diverse and engaging ad texts is crucial for capturing a broad audience and avoiding advertising fatigue. Regardless of the…
FaithCAMERA: Construction of a Faithful Dataset for Ad Text Generation
Akihiko Kato, Masato Mita, Soichiro Murakami +3
In ad text generation (ATG), desirable ad text is both faithful and informative. That is, it should be faithful to the input document, while at the same time containing important i…
Reinforcement Learning for Edit-Based Non-Autoregressive Neural Machine Translation
Hao Wang, Tetsuro Morimura, Ukyo Honda +1
Non-autoregressive (NAR) language models are known for their low latency in neural machine translation (NMT). However, a performance gap exists between NAR and autoregressive model…
Not Eliminate but Aggregate: Post-Hoc Control over Mixture-of-Experts to Address Shortcut Shifts in Natural Language Understanding
Ukyo Honda, Tatsushi Oka, Peinan Zhang +1
Recent models for natural language understanding are inclined to exploit simple patterns in datasets, commonly known as shortcuts. These shortcuts hinge on spurious correlations be…
Annotation-Efficient Language Model Alignment via Diverse and Representative Response Texts
Yuu Jinnai, Ukyo Honda
Preference optimization is a standard approach to fine-tuning large language models to align with human preferences. The quantity, diversity, and representativeness of the preferen…
On the True Distribution Approximation of Minimum Bayes-Risk Decoding
Atsumoto Ohashi, Ukyo Honda, Tetsuro Morimura +1
Minimum Bayes-risk (MBR) decoding has recently gained renewed attention in text generation. MBR decoding considers texts sampled from a model as pseudo-references and selects the t…
A Single Linear Layer Yields Task-Adapted Low-Rank Matrices
Hwichan Kim, Shota Sasaki, Sho Hoshino +1
Low-Rank Adaptation (LoRA) is a widely used Parameter-Efficient Fine-Tuning (PEFT) method that updates an initial weight matrix with a delta matrix consisted by two low-…
Generating Diverse and High-Quality Texts by Minimum Bayes Risk Decoding
Yuu Jinnai, Ukyo Honda, Tetsuro Morimura +1
One of the most important challenges in text generation systems is to produce outputs that are not only correct but also diverse. Recently, Minimum Bayes-Risk (MBR) decoding has ga…
Model-Based Minimum Bayes Risk Decoding for Text Generation
Yuu Jinnai, Tetsuro Morimura, Ukyo Honda +2
Minimum Bayes Risk (MBR) decoding has been shown to be a powerful alternative to beam search decoding in a variety of text generation tasks. MBR decoding selects a hypothesis from…
On the Depth between Beam Search and Exhaustive Search for Text Generation
Yuu Jinnai, Tetsuro Morimura, Ukyo Honda
Beam search and exhaustive search are two extreme ends of text decoding algorithms with respect to the search depth. Beam search is limited in both search width and depth, whereas…
Switching to Discriminative Image Captioning by Relieving a Bottleneck of Reinforcement Learning
Ukyo Honda, Taro Watanabe, Yuji Matsumoto
Discriminativeness is a desirable feature of image captions: captions should describe the characteristic details of input images. However, recent high-performing captioning models,…
Removing Word-Level Spurious Alignment between Images and Pseudo-Captions in Unsupervised Image Captioning
Ukyo Honda, Yoshitaka Ushiku, Atsushi Hashimoto +2
Unsupervised image captioning is a challenging task that aims at generating captions without the supervision of image-sentence pairs, but only with images and sentences drawn from…