The big events that happened this year were graduating from Stanford and starting my PhD at Penn. Graduating from Stanford simply felt like the nominal end to a very long and murky transition, spanning all four years of college, away from the certainty of being a student and into the complete openness of the rest of my life. Retrospectively, I made a lot of progress during college in figuring out what I wanted to do and how I wanted to live, but I certainly felt nowhere close to clarity when I graduated.

But the nice thing is, sometimes, things just happen, even if just yesterday it felt like you had no idea how to get there. Even after I got into grad school this spring, I still wasn’t really certain I wanted to go, but nothing else was more compelling and research was what I had wanted to do for a long time, so I figured I’d at least give it a shot. But luckily, within my first month at Penn, everything just fell into place.

A large part of my college-era uncertainty about my work was that, the more I read papers and worked on projects, the further I felt from being able to even touch the big questions that originally got me interested in these areas, questions which many of papers claim to contribute to but (I felt) made no significant progress towards. But in the second half of the year, at Penn, I started see a path forward, to actually build an understanding of these big questions at the level at which I find meaningful. Once I had this, all my other uncertainties resolved themselves, and I finally felt like I was in the clear and excited to work!

Work (Penn-onwards)

At a high level, I’ve made significant progress in my work, both on some meta-work stuff and in the actual content of my work. For the meta-work stuff, this included creating good scaffolding to do work (e.g. committing to a regular sleep schedule and work hours; setting up a Notion system for planning longer-term goals, organizing notes for different projects, reading and taking notes on papers, and logging daily progress; thinking through how to deal with energy dips and different moods during my work day), which has been pretty crucial for me, given the lack of structure in grad school and my inability to function well without it. I’ve also done good thinking on some more open questions, such as how to develop taste in problems and what I really want to achieve with my work, but of course these are ongoing.

With respect to the content of my work, this can be divided into strategy and specific projects. By strategy, I mean committing to a higher-level goal, then mapping out the pieces needed and how to connect them to build up to a satisfyingly high-level answer. Of course this is something that you never stop adjusting, but I at least now feel like I know where to start. I gave a first pass at articulating my research interests here.

Now onto specific projects. I mainly worked on three somewhat distinct projects. First, I worked on turning my undergrad syntax project into a journal submission. At the end of the semester, this culminated in an article-length rough draft of the paper.

Second, I did an independent reading with Charles. At the beginning of the semester, I put together a syllabus for the reading, focused on the question: How are human language learners similar to and different from other learning systems, specifically artificial neural networks? In the first few weeks of the semester, I was thinking about the induction problem and learning (e.g. most of Language and Learning: The Debate Between Piaget and Chomsky; various Chomsky; some Quine; Berwick 1986; Goodman 1955, 1958, 1983; Lenneberg 2007; Marcus 1993; some Bayesian stuff like Tenenbaum 1999). I read some random ML and statistical learning theory stuff. Then I read some stuff on language acquisition, to get a sense of the relevant developmental facts (Part 1 of O’Grady’s Syntactic Development). It is around this time that I got distracted and veered away from my original reading plan. I came into the semester thinking I would focus for now on computational language learning, but in part due to Don’s class, I started thinking much more about historical linguistics and an acquisition-centered approach to language change.

That leads me to the third project. I took an introduction to historical linguistics class with Don, which naturally made me think about change. I’m using the final paper for that class as the beginning of a project on verb movement in the history of English, which I’ve enjoyed working on the past few weeks. No deliverables yet though.

Work (Stanford)

To avoid being too recency-biased, I’ll write a few notes on what I was working on at Stanford in the first half of the year.

Throughout the year, I was working on a project in Marc’s lab doing some statistical (phylogenetic) analyses of language and genetic data. The paper is basically done but still not submitted due to some hold-ups beyond my control.

During winter quarter, I took CS 229 Machine Learning. For my final project, I worked on designing/training neural networks for molecular property prediction. I continued working on that project sporadically, but two months ago, I finally decided to put it on indefinite hiatus, since I don’t forsee fitting it into my schedule in the near future. I was and still am quite interested in ML for drug discovery and would be interested in working on it more. Maybe once I get more settled into my main research.

In the spring, I took STATS 320 Machine Learning Methods for Neural Data Analysis. It was interesting to implement various models for different tasks in analyzing neural data, and for my final project, I worked on classifying mouse behaviors using autoregressive HMMs (for seizure prediction).

Not work

Work has been the main orientation of (especially the second half of) my year, but there were also a lot of other fun things this year. In this section I’ll talk about some non-work things I’ve been thinking about, reading, watching, or doing.

I did some nice traveling over the summer, in June and July (see posts on China and Japan trips). Before the Japan trip, I read some translated Japanese literature (Kokoro by Natsume Sōseki and Bashō’s The Narrow Road to the Deep North ), as well as a comic rendition of the folklore collection Tono Monogatari. The Narrow Road to the Deep North is a haiku+prose collection on Bashō’s travels in Tohoku, covering lots of the same land we drove through. But honestly the collection didn’t quite work for me: for some writing, the essence comes more from its form rather than its content, and IMO this type of writing rarely translates well. My guess is haiku (similar to basically all Chinese poetry and literary prose) is like that.

From an outsider’s impression of Japan, one aspect of the culture that I always found interesting is that it seems like folklore is still a big part of the present-day cultural imagination. While the kind of interwovenness of mythology into mundane and mostly secular life is not uncommon, in places like China this kind of stuff feels much more distant and marginal in today’s culture. This is in contrast to the Japan that is constructed in, say, Studio Ghibli movies and some anime; just think of all the spirits and gods in Princess Mononoke, My Neighbor Totoro, and Spirited Away. Tono Monogatari fits into this area of interest. I read the comic adaptation by Shigeru Mizuki of Tono Monogatari, which is a collection of local folklore from around Tono, a valley town in central Iwate, documented by Kunio Yanagita and Kizen Sasaki in 1910. We visited Tono on our trip, which was fun. I haven’t really thought enough about these questions around mythology and culture to have anything valuable to say, but at some point I may think more and write something up.

On the topic of Studio Ghibli. I’ve watched a decent number of Ghibli movies before this year (Castle in the Sky, My Neighbor Totoro, Kiki’s Delivery Service, Porco Rosso, Princess Mononoke, Spirited Away, Howl’s Moving Castle, Ponyo, Arrietty, The Boy and the Heron). This year I watched Only Yesterday and The Wind Rises. I really, really liked The Wind Rises: it is now my favorite Ghibli movie and probably my favorite movie in general (though I don’t watch many movies). It is just one of those pieces of art that (to me) is beautiful and feels very true. The first piece in the movie’s soundtrack, “A Journey (A Dream of Flight)”, has also usurped “Merry-Go-Round of Life” as my favorite piece of Ghibli music. I also watched the 2013 documentary The Kingdom of Dreams and Madness on Studio Ghibli, which was pretty good.

In early August, Jasper and I moved to Philadelphia. In the few weeks before my program started, aside from buying furniture and other moving-related tasks, I also started this blog for real. I had already set up the website earlier in the year, but only in August did I start to compile some thoughts and photos from various travels into my first few posts.

Another hobby I picked up around then was tea. I’ve always liked beverages (e.g. espresso-based drinks and matcha), but this year I more systematically got into Chinese tea. I read a random book I found at the library introducing the production and varieties of tea, as well as《茶经述评》, which is a modern commentary on 陆羽《茶经》(Lu Yu’s The Classic of Tea ), the first comprehensive treatise on tea, from the Tang dynasty. I also watched 8 of the 10 episodes in 《茶届中国》, a documentary series on tea produced by Jiangsu TV. Ultimately my goal is just to develop taste, so it is helpful to get a sense of how people categorize and judge tea to build up some guiderails. I’ve also been writing notes on all the tea I drink, but I don’t have good enough taste yet to share those thoughts.

Tangential to my interest in tea, for one day I went on a long rabbithole on the ancient tea-horse road (茶马古道), which is a network of trading routes (from Tang onwards) spanning modern-day Tibet, Sichuan, Yunnan, and Guizhou, and centered around the export of horses from Tibet in exchange for tea from these tea-producing regions of China. Academic interest in the tea-horse road started from an expedition into these areas in the 90s by a team, including the linguist 陈保亚. I watched the few lectures on the tea-horse road from his course on language and culture at Peking, got excited, then spent the rest of the day reading random Baidu Baike pages and looking at lot of maps of Sichuan’s waterways and topography. I started writing some notes for a post on the tea-horse road and Sichuan but ran out of stamina. Maybe later.

Lastly, the other books I’ve read this year: The Roots of Romanticism by Isaiah Berlin; On Language by Noam Chomsky; On China by Henry Kissinger; Zen and the Art of Motorcycle Maintenance by Robert M. Pirsig, a reread of a book I really liked in high school; East of Eden by John Steinbeck; Death of Ivan Ilyich by Leo Tolstoy; 鲁迅《朝花夕拾》and《呐喊》, which I reread upon visiting Sendai; 马伯庸《食南之徒》; and Atomic Habits by James Clear. For the last few months, I’ve also been reading《道德经》(Tao Te Ching), which is more of a longer-term project, since I’m memorizing chunks of it at a time and chewing them over slowly. At some point I may take on a more involved project of writing a translation and commentary of it.

Looking towards next year

In the first part of the year, I hope to get some papers submitted: the journal paper with Marc (if things work out), my syntax journal paper, and a conference abstract on the project on verb movement I just started. I also hope to get some new projects off the ground: a computational project on learning and a statistical project working with the Labov data. These are two big areas of my research interests that I haven’t thought much about this semester, so I’ll need to do some thinking.

I will also continue finetuning parts of my system. I want to better fit in non-work reading and exercise into my routine. I also want to think more about what all I want to dedicate consistent time and brain-space on (e.g. some more non-intellectual, “real” world stuff) and how to incorporate them into my system.

In general, I hope to go full-speed on my work next year, get more clarity about how I want to live my life, and keep enjoying it all! Happy 2026!