Taiming Lu (@TaimingLu) header
Taiming Lu (@TaimingLu) profile picture

Ph.D student at @Princeton | Formerly @JohnsHopkins ’25, @HopkinsDSAI @JHUCompSci @jhuclsp @CCVLatJHU | AI/ML/NLP/CV

Princeton, NJ
Joined June 2024
Photo by Taiming Lu (@TaimingLu)Photo by Taiming Lu (@TaimingLu)Photo by Taiming Lu (@TaimingLu)Photo by Taiming Lu (@TaimingLu)Photo by Taiming Lu (@TaimingLu)Photo by Taiming Lu (@TaimingLu)Photo by Taiming Lu (@TaimingLu)Photo by Taiming Lu (@TaimingLu)Photo by Taiming Lu (@TaimingLu)Photo by Taiming Lu (@TaimingLu)
How do you find the best agentic reasoning strategy for a given task? We introduce Strategist, an agent that evolves better llm reasoning strategies as Python programs, storing each one in a library so the next task can start from it. Yes, chain-of-thought was invented by human, but the next one might not be! Details in thread 🧵
6
8
1
42
6,449
On 30 benchmarks, Strategist beats the strongest hand-written baseline on 24, with the same average accuracy at 39% lower cost. On 10 new tasks, the library alone, with no search, is more accurate than the baseline on 8.
1
1
123
On 30 benchmarks, Strategist beats the strongest hand-written baseline on 24, with the same average accuracy at 39% lower cost. On 10 new tasks, the library alone, with no search, is more accurate than the baseline on 8.
2
136
You want a strong small LLM. Would you start small — or inherit from something bigger? 📄 New paper: Small LLMs: Pruning vs. Training from Scratch We find that pruning is more than a better initialization: simply giving randomly initialized LLMs more training tokens is often not enough to catch up.
14
119
10
617
54,146
Taiming Lu retweeted
Can a small academic team build a strong text-to-image model using only public datasets? Introducing i1: a simple, fully open recipe for strong text-to-image models
11
78
9
442
71,766
Knowledge doesn't always flow downhill. We find that in LLM pretraining, a weaker teacher can improve a stronger student, and pushing the teacher further can actually hurt. New paper: Strong Teacher Not Needed? On Distillation in LLM Pretraining.
7
57
6
348
48,799
Distillation improves generalization more readily than in-domain fit. Out-of-distribution perplexity and downstream accuracy improve more consistently than in-domain perplexity, where some configurations help OOD/downstream while doing nothing for in-domain.
1
2
7
1,234
Teacher–student compatibility matters more than raw teacher strength. This changes how you pick a teacher: both for frontier training (where the best available teacher is often a prior generation) and for efficient small models, where "bigger teacher is better" isn't the right rule. Thanks @liuzhuang1234 for the support! arxiv: arxiv.org/abs/2605.23857 code: github.com/zlab-princeton/st…
1
2
10
1,155