Blog Post··1 min read
Notes on Maintaining a Research Blog
A lightweight example post showing the informal article path for the site.
AGI Research Scientist
Swastik Roy
Research notes on AI systems, evaluation, and alignment
Swastik Roy works on AI systems, evaluation, alignment, post-training, and reliable measurement. Replace this placeholder with an author-approved biography.
Research Interests
- LLM evaluation and model behavior
- Model specifications, AI constitutions, and alignment
- Post-training, reward models, rubrics, judges, and RL environments
- Reliable benchmark and evaluation methodology
Recent Articles
All articlesBlog Post··4 min read
Activation Steering: A New Frontier in AI Control—But Does It Scale?
A practical introduction to activation steering, why it is promising for AI control, and where scaling challenges still remain.
Selected Research
Research archiveresearch-project
Reliable LLM Evaluation Methods
Placeholder for author-reviewed research on rubrics, judges, benchmark design, and reproducible evaluation methodology.
research-project
Model Behavior and Specification Studies
Placeholder for work on model specifications, constitutions, behavioral audits, and alignment-relevant post-training analysis.
Selected Publications
JudgeBoard: Benchmarking and Enhancing Small Language Models for Reasoning Evaluation
Zhenyu Bi, Gaurav Srivastava, Yang Li, Swastik Roy, Meng Lu, Morteza Ziyadi, Xuan Wang · Proceedings of the AAAI Conference on Artificial Intelligence · 2026
Publication sourceTalks
Talk metadata needed
Placeholder · 2026-06-16
Newsletter integration placeholder. Add a provider endpoint when ready.