Accepted · 2026
DoubtBench: Does Self-Doubt Help Language Models—or Just Make Them Check? New
Findings of WSDM 2027
Separating the effect of self-doubt wording from direct instructions to check.
Research
Benchmarks and papers on evaluating AI systems against the claims they actually make.
Accepted · 2026
Do LLM Reviewers Respect Scope? ScopeBench-PR: A Benchmark for Scope Fairness in Peer Review
GlobalSouthAI Workshop · IJCAI–ECAI 2026
Awarded Best Paper Presentation
Runner-up · Three Minute Thesis (3MT)
Measuring whether LLM reviewers respect a paper’s stated scope.
Published · 2026
Retrieval Mechanisms Surpass Long-Context Scaling in Time Series Forecasting
ICLR 2026 · TSALM Workshop
Awarded ICLR 2026 grant · $2,025
Selective retrieval for time-series forecasting beyond long-context scaling.