Ranking Reasoning LLMs under Test-Time ScalingACL 2026 presentation on ranking reasoning LLMs under test-time scaling: dense repeated-trial evaluation, Bayes@N as a practical default, low-budget priors, categorical ranking, and the Scorio toolkit.
Serving reasoning LLMs efficiently and reliably: lossless DFloat11 compression, KV-cache quantization, and Bayes@N evaluation and ranking under test-time scaling.
Serving reasoning LLMs efficiently and reliably: lossless DFloat11 compression, KV-cache quantization, and Bayes@N evaluation and ranking under test-time scaling.
Quantize What Counts: More for Keys, Less for ValuesACL 2026 presentation on Quantize What Counts: More for Keys, Less for Values, explaining key-value norm disparity, key-prioritized quantization, and practical KV-cache compression guidance.
Python environments, how to create and reproduce them, and when to use pip, conda, micromamba, uv, pipx, lockfiles, and containers.
Don't Pass@k: A Bayesian Framework for LLM EvaluationICLR 2026 presentation on Don't Pass@k: a Bayesian evaluation framework (Bayes@N) with Dirichlet posteriors, credible intervals, a non-overlap decision rule, categorical rubric scoring, and the Scorio toolkit.
SCIPE Workshop on Large Language Models • Final Presentation
SCIPE Workshop on LLMs - Day 3
SCIPE Workshop on LLMs - Day 2
SCIPE Workshop on LLMs
70% Size, 100% Accuracy: Lossless LLM Compression for Efficient GPU Inference via Dynamic-Length FloatNeurIPS 2025 presentation on Dynamic-Length Float (DFloat11/DF11): a lossless format that Huffman-codes BFloat16 exponents down to ~11 bits, cutting model size ~30% with bit-for-bit identical outputs and a GPU kernel that makes compressed inference fast.
M.Sc. Thesis in Computer Science
A principled Bayesian framework that replaces Pass@k with posterior estimates, credible intervals, and stable rankings for LLM evaluation