7 papers · 1 filter
Tokenization with Split Trees
Craig W. Schmidt, Michael Krumdick, Adam Wiemerslage +4
We introduce Tokenization with Split Trees (ToaST), a subword tokenization method that directly optimizes compression under a new recursive inference procedure. ToaST greedily spli…
FrontierFinance: A Long-Horizon Computer-Use Benchmark of Real-World Financial Tasks
Michael Krumdick, Varshini Reddy, Shivani Chaudhary +8
As concerns surrounding AI-driven labor displacement intensify in knowledge-intensive sectors, existing benchmarks fail to measure performance on tasks that define practical profes…
Cost-Efficient Estimation of General Abilities Across Benchmarks
Michael Krumdick, Adam Wiemerslage, Seth Ebner +2
Thousands of diverse benchmarks have been developed to measure the quality of large language models (LLMs). Yet prior work has demonstrated that LLM performance is often sufficient…
No Free Labels: Limitations of LLM-as-a-Judge Without Human Grounding
Michael Krumdick, Charles Lovering, Varshini Reddy +2
Reliable evaluation of large language models (LLMs) is critical as their deployment rapidly expands, particularly in high-stakes domains such as business and finance. The LLM-as-a-…
On Finding Inconsistencies in Documents
Charles J. Lovering, Seth Ebner, Brandon Smock +5
Professionals in academia, law, and finance audit their documents because inconsistencies can result in monetary, reputational, and scientific costs. Language models (LMs) have the…
BLEUBERI: BLEU is a surprisingly effective reward for instruction following
Yapei Chang, Yekyung Kim, Michael Krumdick +4
Reward models are central to aligning LLMs with human preferences, but they are costly to train, requiring large-scale human-labeled preference data and powerful pretrained LLM bac…