ICLR 2025 accepted works include 'Hot-pluggable Federated Learning', 'STBLLM: Breaking the 1-Bit Barrier with Structured Binary LLMs', and two blogposts on the Lottery LLM Hypothesis and reflections on LLM-based human simulations.
AAAI 2025 Oral paper 'ParZC: Parametric Zero-Cost Proxies for Efficient NAS' introduces a novel zero-cost proxy method for neural architecture search.
Named Top Reviewer of NeurIPS 2024 for both main and D&B tracks.
Paper 'Hot Pluggable Federated Learning' received the Outstanding Student Paper Award at FL@FM-NeurIPS’24 Workshop.