Mark J. Nelson

Small-scale study: 16 law professors were asked to judge short-answer Q&A practice materials written by one of the other 15 professors, or generated by an LLM (Gemini 2.5 Pro). They preferred the LLM materials in 75% of cases, with fairly strong inter-rater agreement. Has some…

law.stanford.edu

An interesting development on the local-models front: one of the largest U.S. law firms is going to start fine-tuning and hosting its own LLMs in-house.

ft.com

finally some off-the-grid local AI

CrankGPT — fully offline, human-powered local AI squeezlabs.github.io

did you see this one a few months ago?

China’s AI Boyfriend Business Is Taking On a Life of Its Own wired.com

paper:

Language Model Crossover: Variation through Few-Shot Prompting arxiv.org

Kind of reassuring to read that one of the biggest current problems in ML for drug discovery isn't any kind of exotic new AI/ML problem but just, still, the difficulty of preventing data leakage from the test set.

science.org

Dealing with too many submissions to TMLR by deriving "a Generalized Harmonic Quota Rule, a framework that subsumes the Harmonic Quota Rule and other natural quota rules" in a 12-page paper is a very The Machine Learning Community solution to this particular problem.

How Many Submissions May an Author Make? A Harmonic Quota for Submissions under Coauthorship arxiv.org

Open-Rank Professor of Interactive Media and Games – Game Art and Animation, at the University of Southern California. Listed under Preferred Experience: "Familiarity with emerging AI-assisted art and content creation workflows" 👀

Open-Rank Professor of Interactive Media and Games - Game Art and Animation at USC usccareers.usc.edu

Underreported thing about Gemini (more than other LLMs I think) is that it's an ok replacement for Google Books, and fluidly multilingual. Like I can ask a question and request answers be based only on books by a specific academic (which are in Greek) and it will dig up relevant passages.

I like AI, but still, is this what I want to see at the Amtrak station

Noticing a pattern across departments that faculty are generally a lot more excited about AI than students, esp undergrad students.

Got a good 96gb-VRAM local LLM harness going and can't believe I used to just... put up with the garbage latencies that OpenAI/Anthropic/Google serve up.

There's a huge increase in single-author papers being submitted by early-career researchers in AI/ML. Leaving aside other issues, from a purely CV-maxxing perspective do these even help? I think on a hiring committee I'd give them mixed to negative weight in many cases.

A reason I've pulled back from reviewing for big AI conferences is a feeling that I'm doing unpaid supervision of other people's PhD students. Too many ppl submitting 10+ papers to a single conference where I doubt the prof whose name is on the paper has done a thorough review & revision themselves.

Not a strongly held opinion, but I'm a little skeptical of the recent LLM math results not really being compared against baseline search methods with similar compute budgets. Some of them are using pretty huge compute budgets!

An interesting thing about LLMs in Python is that they seem to broadly push code towards some kind of conventional wisdom about best practices, as judged maybe by whoever is setting up the posttraining recipes (I say "in Python" mostly because I notice that more strongly in Python).

Tip for people stuck in the Microsoft 365 ecosystem: the Copilot AI search sidebar is a huge improvement! Largely because almost anything is better than Outlook search (it won't find emails that have the verbatim search term in the subject!). But still, now there's working search. Ty AI boom.

A thing LLM coding sort of makes more feasible is making *smaller* personalized apps. Like instead of WMATA's big and annoying to use transit app, I'm trying out a custom little thing that just shows me the 2 bus lines I take. Lines hardcoded; stops hardcoded; no configuration; barely any interface.