Things to Do in L.A. Tap to enable a layout that focuses on the article. AI pilot program in L.A. County courts will help judges craft rulings in some cases This is read by an automated voice. Please report any issues or inconsistencies here . A select panel of L.A. County judges now have access to an artificial intelligence tool that can help them summarize motions and draft rulings in civil court. The tool, Learned Hand, is already in use by judges in 10 states, according to the company's CEO.

Recent work has shown that language models' (LMs) prompt-based learning capabilities make them well suited for automating data labeling in domains where manual annotation is expensive.

Recent work has shown that language models' (LMs) prompt-based learning capabilities make them well suited for automating data labeling in domains where manual annotation is expensive. The challenge is that while writing an initial prompt is cheap, improving a prompt is costly -- practitioners often require significant labeled data in order to evaluate the impact of prompt modifications. Our work asks whether it is possible to improve prompt-based learning without additional labeled data. We approach this problem by attempting to modify the predictions of a prompt, rather than the prompt itself. Our intuition is that accurate predictions should also be consistent: samples which are similar under some feature representation should receive the same prompt prediction. We propose Embroid, a method which computes multiple representations of a dataset under different embedding functions, and uses the consistency between the LM predictions for neighboring samples to identify mispredictions. Embroid then uses these neighborhoods to create additional predictions for each sample, and combines these predictions with a simple latent variable graphical model in order to generate a final corrected prediction. In addition to providing a theoretical analysis of Embroid, we conduct a rigorous empirical evaluation across six different LMs and up to 95 different tasks. We find that (1) Embroid substantially improves performance over original prompts (e.g., by an average of 7.3 points on GPT-JT), (2) also realizes improvements for more sophisticated prompting strategies (e.g., chain-of-thought), and (3) can be specialized to domains like law through the embedding functions.

When looking for answers to legal questions, people increasingly start their searches online. But what they find isn't always very useful--prompting the law schools at Stanford University and Suffolk University to team up to harness artificial intelligence (AI) to help people identify their specific legal issues. Historically, machines have struggled to understand context in human speech. For example, if someone says, "I'm getting kicked out of my house," most people understand that the person is not being physically kicked but is rather being removed from his or her home--or, to use the legal term, evicted. But machines typically can't understand "kicked out of my house" as "evicted" without being trained through a large number of similar questions.