From research questions to models, benchmarks, and working tools.
Research into practice
01Apodex · May 2026–present
Built pretraining data infrastructure from scratch and scaled it to web scale, with data quality assessment guiding what is collected.
An acquisition pipeline covering raw HTML, PDFs, and other information-rich documents, with quality scoring to guide collection.
02Apodex · May 2025–May 2026
Built the agent harness from scratch and led its core architecture as development grew to four or five contributors. Designed multi-source verification, parallel tool use, and multi-agent coordination.
A model and workbench for complex research, alongside the open-source FrontierAgent harness.
RESEARCH & TOOL USE
03I²R, A*STAR · 2023–2025
Led data preparation and evaluation, and co-led model training within a six-person team at A*STAR I²R.
MERaLiON AudioLLM, presented at ACL 2025 System Demonstrations, as part of Singapore’s National Multimodal LLM Programme.
AUDIO-LANGUAGE MODELS
04I²R, A*STAR · NAACL 2025
Led AudioBench: eight tasks and 26 datasets for evaluating audio-language models. First author.
A universal audio-language benchmark, published at NAACL 2025.
AUDIO MODEL EVALUATION
05I²R, A*STAR · NAACL 2024
Led SeaEval to evaluate cross-lingual consistency and cultural reasoning. Co-first author.
A multilingual evaluation benchmark, published at NAACL 2024.
ACROSS LANGUAGES & CULTURES
Models and frameworks I worked on as part of a team.
Resources I maintain and experiments I have built.