Profiles2022-2025
Dongliang Xu
Du Xiaoman (Beijing) Science Technology Co., Ltd.
12Papers
3Last author
39Unique co-authors
0First author
0Solo
First and last author come from byline position, which some venues assign alphabetically.
Recent papers12
- Turning Trash into Treasure: Accelerating Inference of Large Language Models with Token RecyclingAuthor 7 of 7ACL 2025 · 31 citations
- Beyond Similarity: A Gradient-based Graph Method for Instruction Tuning Data SelectionAuthor 9 of 13ACL 2025 · 3 citations
- Length Extrapolation of Transformers: A Survey from the Perspective of Positional EncodingAuthor 5 of 9EMNLP Findings 2024 · 14 citations
- GlobeSumm: A Challenging Benchmark Towards Unifying Multi-lingual, Cross-lingual and Multi-document News SummarizationAuthor 6 of 10EMNLP 2024 · 8 citations
- Advancing Large Language Model Attribution through Self-ImprovingAuthor 7 of 10EMNLP 2024 · 2 citations
- Extending Context Window of Large Language Models from a Distributional PerspectiveAuthor 5 of 8EMNLP 2024 · 1 citation
- How does Architecture Influence the Base Capabilities of Pre-trained Language Models? A Case Study Based on FFN-Wider and MoE TransformersAuthor 6 of 6NeurIPS 2024 · 1 citation
- Python is Not Always the Best Choice: Embracing Multilingual Program of ThoughtsAuthor 7 of 8EMNLP 2024 · 1 citation
- Make Some Noise: Unlocking Language Model Parallel Inference Capability through Noisy TrainingAuthor 8 of 9EMNLP 2024
- Meaningful Learning: Enhancing Abstract Reasoning in Large Language Models via Generic Fact GuidanceAuthor 5 of 8NeurIPS 2024
- TranS: Transition-based Knowledge Graph Embedding with Synthetic Relation RepresentationAuthor 3 of 3EMNLP Findings 2022 · 14 citations
- Instance-Guided Prompt Learning for Few-Shot Text MatchingAuthor 7 of 8EMNLP Findings 2022 · 1 citation
