Profiles2025-2026
Xianzhi Yu
Huawei Advanced Computing and Storage Lab
5Papers
1Last author
23Unique co-authors
0First author
0Solo
First and last author come from byline position, which some venues assign alphabetically.
Recent papers5
- TrimR: Verifier-based Training-Free Thinking Trimming for Efficient Test-Time ScalingAuthor 7 of 10ICLR 2026 · 14 citations
- Benchmarking Post-Training Quantization of Large Language Models under Microscaling Floating Point FormatsAuthor 7 of 7ACL 2026 · 2 citations
- MoESD: Unveil Speculative Decoding's Potential for Accelerating Sparse MoEAuthor 6 of 8NeurIPS 2025 · 23 citations
- AttentionPredictor: Temporal Patterns Matter for KV Cache CompressionAuthor 7 of 11NeurIPS 2025 · 8 citations
- Faster and Better LLMs via Latency-Aware Test-Time ScalingAuthor 5 of 8EMNLP Findings 2025 · 1 citation
