⚡ [Performance] Vectorize adjacency matrix construction in SymbolicDataset

[Vishal-sys-code]

💡 What: Replaced the $O(N^2)$ nested Python loops used to construct the A_rel adjacency matrix with highly efficient, vectorized PyTorch operations (using torch.isin and dimension broadcasting).
🎯 Why: The previous nested loop was a performance bottleneck during synthetic dataset generation, consuming roughly 0.43 seconds per 100 samples alone.
📊 Measured Improvement: In isolated benchmarks using a sequence length of 100 over 100 iterations, the raw array operation time improved from 0.4325s to 0.0102s (a ~42x speedup). Overall single-batch dataset initialization (2000 samples) dropped from 0.57s to 0.32s, effectively cutting the generation time by ~44% without any changes in functionality.

---

PR created automatically by Jules for task 3777852949880920115 started by @Vishal-sys-code

[google-labs-jules]

👋 Jules, reporting for duty! I'm here to lend a hand with this pull request.

When you start a review, I'll add a 👀 emoji to each comment to let you know I've read it. I'll focus on feedback directed at me and will do my best to stay out of conversations between you and other bots or reviewers to keep the noise down.

I'll push a commit with your requested changes shortly after. Please note there might be a delay between these steps, but rest assured I'm on the job!

For more direct control, you can switch me to Reactive Mode. When this mode is on, I will only act on comments where you specifically mention me with @jules. You can find this option in the Pull Request section of your global Jules UI settings. You can always switch back!

New to Jules? Learn more at jules.google/docs.

---

For security, I will only act on instructions from the user who triggered this task.

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
variational-linear-attention	Ready	Preview, Comment	Apr 8, 2026 6:41am