4 citations · 5 across the 10 of their papers we have counts for
6 papers · 1 filter
SQ-format: A Unified Sparse-Quantized Hardware-friendly Data Format for LLMs
Ruixuan Huang, Hao Zeng, Hantao Huang +4
Post-training quantization (PTQ) plays a crucial role in the democratization of large language models (LLMs). However, existing low-bit quantization and sparsification techniques a…
A note on conditional PAC-efficient reasoning in large language model routing
Hao Zeng, Bingyi Jing
We study distribution-free risk control for model routing, motivated by large language model reasoning. We formalize pointwise conditional efficiency under a probably approximately…
Identity-Compatible Auctions
Haoyuan Zeng
This paper studies sellers' and buyers' incentives to engage in shill bidding in single-object auctions. An auction is seller identity-compatible if the seller cannot profit by pos…
Persuasive Selection in Signaling Games
Haoyuan Zeng
This paper introduces a novel criterion, persuasiveness, to select equilibria in signaling games. In response to the Stiglitz critique, persuasiveness focuses on the comparison acr…
Mechanism Design with Information Leakage
Samuel Häfner, Marek Pycia, Haoyuan Zeng
We study the design of mechanisms---e.g., auctions---when the designer does not control information flows between mechanism participants. A mechanism equilibrium is leakage-proof i…
GLM-4.5: Agentic, Reasoning, and Coding (ARC) Foundation Models
5 Team, Aohan Zeng, Xin Lv +167
We present GLM-4.5, an open-source Mixture-of-Experts (MoE) large language model with 355B total parameters and 32B activated parameters, featuring a hybrid reasoning method that s…