Dustin Wright

Which, whose, and how much knowledge do LLMs represent? I'm excited to share our preprint answering these questions: "Epistemic Diversity and Knowledge Collapse in Large Language Models" 📄Paper: arxiv.org/pdf/2510.04226 💻Code: github.com/dwright37/ll... 1/10

🛣️Methodology can be used in the future to study epistemic diversity for any arbitrary topics, downstream tasks, and real-world use cases with open-ended plain-text LLM outputs. This allows researchers to answer research questions about which, whose, and how much knowledge LLMs are representing 9/10

Which, whose, and how much knowledge do LLMs represent? I'm excited to share our preprint answering these questions: "Epistemic Diversity and Knowledge Collapse in Large Language Models" 📄Paper: arxiv.org/pdf/2510.04226 💻Code: github.com/dwright37/ll... 1/10

🪛 Approach: we propose a new methodology which includes sampling plain text LLM outputs with 200 prompt variations from real chats across 155 topics, decomposing into individual claims, and clustering those claims based on entailment. 7/10

Which, whose, and how much knowledge do LLMs represent? I'm excited to share our preprint answering these questions: "Epistemic Diversity and Knowledge Collapse in Large Language Models" 📄Paper: arxiv.org/pdf/2510.04226 💻Code: github.com/dwright37/ll... 1/10

🏗️ Model size has an unintuitive negative impact on diversity; smaller models tend to be more diverse 🔎 RAG has a positive impact on diversity, indicating its usefulness in making LLM outputs more diverse. However, the gains from RAG are not equal across topics about different countries 5/10

Which, whose, and how much knowledge do LLMs represent? I'm excited to share our preprint answering these questions: "Epistemic Diversity and Knowledge Collapse in Large Language Models" 📄Paper: arxiv.org/pdf/2510.04226 💻Code: github.com/dwright37/ll... 1/10

👍 To assess this risk, we set out to measure to what extent LLMs are homogenous in terms of the *real-world claims* they generate. We perform a large study across 27 LLMs, 2 generation settings, with different model versions and sizes. In a nutshell, our findings are: 3/10

Which, whose, and how much knowledge do LLMs represent? I'm excited to share our preprint answering these questions: "Epistemic Diversity and Knowledge Collapse in Large Language Models" 📄Paper: arxiv.org/pdf/2510.04226 💻Code: github.com/dwright37/ll... 1/10

🎉 Our work on attribution in summarization is now accepted to #EMNLP2025 main! 🎉 "Unstructured Evidence Attribution for Long Context Query Focused Summarization" w/ @zainmujahid.me , Lu Wang, @iaugenstein.bsky.social , and @davidjurgens.bsky.social

🦾 We demonstrate across 5 LLMs and 4 datasets that LLMs adapted with SUnsET generate more relevant and factually consistent evidence, extract evidence from more diverse locations in their context, and can generate more relevant and consistent summaries than baselines.

🎉 Our work on attribution in summarization is now accepted to #EMNLP2025 main! 🎉 "Unstructured Evidence Attribution for Long Context Query Focused Summarization" w/ @zainmujahid.me , Lu Wang, @iaugenstein.bsky.social , and @davidjurgens.bsky.social

💡 Normally when automatically generated summaries cite supporting evidence, they cite fixed-granular evidence e.g., individual sentences or whole documents. Our work proposes to extract spans of *any* length as more relevant and consistent evidence for long context query focused summaries.

New preprint: "Unstructured Evidence Attribution for Long Context Query Focused Summarization" We propose *unstructured* evidence attribution for long context summarization and a synthetic dataset called SUnsET which can help models perform this task. Paper link: arxiv.org/abs/2502.14409 Thread ⬇️

3) Evidence tends to be lost-in-the-middle for all base models 4) Shuffling the document sections helps mitigate evidence being lost-in-the-middle 5) Learning to cite with SUnsET also improves the quality of the final summaries

New preprint: "Unstructured Evidence Attribution for Long Context Query Focused Summarization" We propose *unstructured* evidence attribution for long context summarization and a synthetic dataset called SUnsET which can help models perform this task. Paper link: arxiv.org/abs/2502.14409 Thread ⬇️

We use SUnsET to adapt LLMs to the unstructured evidence attribution tasks using two approaches: standard fine-tuning and fine-tuning on shuffled documents, in order to overcome the lost-in-the-middle problem for evidence attibution.

New preprint: "Unstructured Evidence Attribution for Long Context Query Focused Summarization" We propose *unstructured* evidence attribution for long context summarization and a synthetic dataset called SUnsET which can help models perform this task. Paper link: arxiv.org/abs/2502.14409 Thread ⬇️

We want to be able to adapt LLMs to this task. The ideal dataset would be extremely difficult to curate by hand: long documents paired with questions, summaries, and free text evidence from the context. To get around this, we create ☀️SUnsET☀️: The Summaries with Unstructured Evidence Text Dataset

New preprint: "Unstructured Evidence Attribution for Long Context Query Focused Summarization" We propose *unstructured* evidence attribution for long context summarization and a synthetic dataset called SUnsET which can help models perform this task. Paper link: arxiv.org/abs/2502.14409 Thread ⬇️

📄 New preprint: "Collecting Cost-Effective, High-Quality Truthfulness Assessments with LLM Summarized Evidence" We show: fact checking w/ crowd workers is more efficient when using LLM summaries, quality doesn't suffer. arxiv.org/abs/2501.18265