4 papers
AsyncLane: Decoupling Refinement from Advancement in Diffusion Language Model Decoding
Yingxuan Ren, Yuxuan Lou, Yong Liu +4
Block-wise semi-autoregressive decoding is the standard inference paradigm for diffusion large language models (DLMs), but it imposes a strict dependency between blocks: the next b…
TOKI: A Bitemporal Operator Algebra for Contradiction Resolution in LLM-Agent Persistent Memory
Ziming Wang
Persistent memory for an LLM agent is a write-heavy substrate: every belief update is a versioned write, and a new claim may contradict a stored one. Production systems use four re…
Fine-Tuned LLMs Know They Don't Know: A Parameter-Efficient Approach to Recovering Honesty
Zeyu Shi, Ziming Wang, Tianyu Chen +4
The honesty of Large Language Models (LLMs) is increasingly important for safe deployment in high-stakes domains. However, this crucial trait is severely undermined by supervised f…
Towards Objective Fine-tuning: How LLMs' Prior Knowledge Causes Potential Poor Calibration?
Ziming Wang, Zeyu Shi, Haoyi Zhou +3
Fine-tuned Large Language Models (LLMs) often demonstrate poor calibration, with their confidence scores misaligned with actual performance. While calibration has been extensively…