Papers by Terufumi Morishita
Mitigating Catastrophic Forgetting in Target Language Adaptation of LLMs via Source-Shielded Updates (2026.acl-long)
Copied to clipboard
| Challenge: | Large language models underperform in languages absent or underrepresented in training data, creating barrier to equitable access for speakers worldwide. |
| Approach: | They propose a selective parameter update strategy that proactively preserves source knowledge by identifying critical parameters critical to maintaining source abilities. |
| Outcome: | Experiments in five typologically diverse languages show that SSU mitigates catastrophic forgetting. |
Disentangling the Effects of Unlearning in Measuring Parametric Faithfulness of Chain-of-Thought (2026.acl-srw)
Copied to clipboard
Ryo Mitsuhashi, Gaku Morio, Ayana Niwa, Masahiro Kaneko, Kentaro Inui, Terufumi Morishita, Yuta Koreeda, Yasuhiro Sogawa
| Challenge: | Chain-of-Thought (CoT) has been debated as a model's faithfulness to internal reasoning process. |
| Approach: | They propose to use unlearning to measure parametric faithfulness of models by adjusting for unintended artifacts of unlearning. |
| Outcome: | The proposed metric accounts for the unintended artifacts of unlearning and shows that it is non-negligible. |
How do different tokenizers perform on downstream tasks in scriptio continua languages?: A case study in Japanese (2023.acl-srw)
Copied to clipboard
| Challenge: | Existing studies on scriptio continua languages lack comprehensiveness of tokenizers . authors use Byte-Pair-Encoding or Unigram instead of WordPiece for subword tokenizer . |
| Approach: | They investigate the effect of tokenizers on the downstream performance of pretrained language models in scriptio continua languages where no explicit spaces exist between words. |
| Outcome: | The proposed tokenizers perform better on a wide range of tasks compared with other tokenizer methods . the results show that each task has an optimal morphological analyzer . |
How does the task complexity of masked pretraining objectives affect downstream performance? (2023.findings-acl)
Copied to clipboard
| Challenge: | Masked language modeling (MLM) is a widely used self-supervised pretraining objective. |
| Approach: | They propose to use a mask-based objective to predict a token that is replaced with a masked token given its context. |
| Outcome: | The proposed objectives show that they should have half the complexity needed to perform comparably to MLM. |
JFLD: A Japanese Benchmark for Deductive Reasoning Based on Formal Logic (2024.lrec-main)
Copied to clipboard
| Challenge: | Large language models (LLMs) have proficiently solved a broad range of tasks with their rich knowledge but struggle with logical reasoning. |
| Approach: | They propose a deductive reasoning benchmark for Japanese that assesses logical reasoning abilities isolated from knowledge and various reasoning rules. |
| Outcome: | The proposed benchmarks assess whether LLMs can generate logical steps to (dis)prove a given hypothesis based on a set of facts. |
Towards Better Non-Tree Argument Mining: Proposition-Level Biaffine Parsing with Task-Specific Parameterization (2020.acl-main)
Copied to clipboard
| Challenge: | Argument mining studies have advanced the ability to predict argument structures, but the technology for capturing non-tree-structured arguments is still in its infancy. |
| Approach: | They propose a neural model that can predict proposition types and edges between propositions. |
| Outcome: | The proposed model improves edge prediction performance compared to baseline models. |
Is Micro Domain-Adaptive Pre-Training Effective for Real-World Operations? Multi-Step Evaluation Reveals Potential and Bottlenecks (2026.eacl-industry)
Copied to clipboard
Masaya Tsunokake, Yuta Koreeda, Terufumi Morishita, Koichi Nagatsuka, Hikaru Tomonari, Yasuhiro Sogawa
| Challenge: | Domain-adaptive pre-training (DAPT) is one approach for enabling LLMs to handle unseen knowledge. |
| Approach: | They propose to disentangle the answering process into three subtasks and evaluate the performance of each subtask. |
| Outcome: | The proposed model resolves the elicitation task that the base model struggled with but does not resolve other subtasks. |
End-to-end Argument Mining with Cross-corpora Multi-task Learning (2022.tacl-1)
Copied to clipboard
| Challenge: | Argument(ation) mining is a task of identifying argument structure from text . lack of training data makes it difficult to train models based on limited data sets. |
| Approach: | They propose an end-to-end cross-corpus argument mining method that uses auxiliary argument mining corpora to train models. |
| Outcome: | The proposed method outperforms models trained on a single corpus on arguments on arguments in argument mining tasks. |
Project-then-Transfer: Effective Two-stage Cross-lingual Transfer for Semantic Dependency Parsing (2021.eacl-main)
Copied to clipboard
| Challenge: | Several remarkable contributions have been made in syntactic dependency parsing, especially on universal dependencies. |
| Approach: | They propose to capture cross-linguality by combing annotation projection and model transfer of pre-trained language models. |
| Outcome: | The proposed model parser almost achieved the approximated upper bound. |