Papers by Kexun Zhang
Scaling LLM Inference Efficiently with Optimized Sample Compute Allocation (2025.naacl-long)
Copied to clipboard
| Challenge: | Existing methods to optimize sample allocations for large language models fail to account for the optimal sampling configuration. |
| Approach: | They propose an algorithm that optimizes sample allocation by finding an optimal mix of different inference configurations. |
| Outcome: | The proposed algorithm achieves better accuracy on SWE-Bench with 3x less compute than the default configuration. |
Human Bias in the Face of AI: Examining Human Judgment Against Text Labeled as AI Generated (2025.findings-acl)
Copied to clipboard
| Challenge: | Prior research on AI mistrust focused primarily on AI's bias towards different human pop-ups. |
| Approach: | They examine how bias shapes the perception of AI versus human generated content . they found that raters favored content labeled "Human Generated" even when labels were deliberately swapped . |
| Outcome: | The findings highlight the limitations of human judgment in interacting with AI and offer a foundation for improving human-AI collaboration. |
Revealing the Barriers of Language Agents in Planning (2025.naacl-long)
Copied to clipboard
| Challenge: | Existing studies show language agents lack human-level planning abilities . limitations and mechanisms to address them remain insufficiently understood . |
| Approach: | They apply a feature attribution study to identify key factors hindering agent planning . they identify the limited role of constraints and diminishing influence of questions . |
| Outcome: | The proposed model achieves 15.6% on a real-world planning benchmark. |
Hire a Linguist!: Learning Endangered Languages in LLMs with In-Context Linguistic Descriptions (2024.findings-acl)
Copied to clipboard
| Challenge: | Existing LLMs rarely perform well in unseen, endangered languages . Existing models such as Llama and GPT-4 lack a rich corpus of training data . |
| Approach: | They propose a training-free approach to enable an LLM to process unseen languages that hardly occur in its pre-training. |
| Outcome: | The proposed approach elevates translation capability from GPT-4’s 0 to 10.5 BLEU for 10 language directions. |
Programming by Example meets Historical Linguistics: A Large Language Model Based Approach to Sound Law Induction (2025.acl-long)
Copied to clipboard
Atharva Naik, Darsh Agrawal, Hong Sng, Clayton Marr, Kexun Zhang, Nathaniel Romney Robinson, Kalvin Chang, Rebecca Byrnes, Aravind Mysore, Carolyn Rose, David R. Mortensen
| Challenge: | Historical linguists have written programs that convert reconstructed words into their attested descendants via ordered string rewrite functions. |
| Approach: | They propose to use a model to generate a "similar distribution" for sound law induction . they propose four kinds of methods with varying amounts of inductive bias to investigate best performance . |
| Outcome: | The proposed model shows that it can be fine tuned with training data and evaluation data. |
Focus on the Action: Learning to Highlight and Summarize Jointly for Email To-Do Items Summarization (2022.findings-acl)
Copied to clipboard
| Challenge: | Existing to-do item generation models focus on generating action mentions to provide more structured summaries of email text. |
| Approach: | They propose a learning to highlight and summarize framework to learn to identify the most salient text and actions and incorporate these structured representations to generate more faithful to-do items. |
| Outcome: | The proposed model outperforms baseline models and achieves state-of-the-art performance in terms of evaluation and human judgement. |
A Study of Syntactic Multi-Modality in Non-Autoregressive Machine Translation (2022.naacl-main)
Copied to clipboard
| Challenge: | Non-autoregressive translation models suffer from the multi-modality problem when a source sentence corresponds to multiple correct translations. |
| Approach: | They propose to decompose the syntactic multi-modality problem into short- and long-range models and evaluate them on synthesized and real datasets. |
| Outcome: | The proposed loss functions can handle short- and long-range syntactic multi-modalities better than existing models. |
Large Language Models Are Partially Primed in Pronoun Interpretation (2023.findings-acl)
Copied to clipboard
| Challenge: | Existing studies suggest large language models acquire rich linguistic representations, but little is known about whether they adapt to linguistic biases in a human-like way. |
| Approach: | They examine whether large language models display human-like referential biases using stimuli and procedures from real psycholinguistic experiments. |
| Outcome: | The proposed models display human-like referential biases when exposed to referential patterns in the local context. |
Extrapolating to Unknown Opinions Using LLMs (2025.coling-main)
Copied to clipboard
| Challenge: | ice cream flavors and climate change are among the topics people hold on various topics. |
| Approach: | They propose to use a large language model to extrapolate from stances to unknown opinions by prompting and fine-tuning data to improve their ability to extrapole from known to unknown stance. |
| Outcome: | The proposed model can extrapolate from opinions on known topics to unknown ones and generate reasoning behind extrapolation. |