4 papers · 1 filter
DriftGuard: Safety-Aware Multi-Monitor Detection and Selective Adaptation for Evolving Toxicity Moderation
Yuting Xin, Hanyu Cai, Binqi Shen +2
Automated toxicity moderation systems operate in dynamic online environments where harmful behavior evolves through coded language, shifting targets, and strategic adaptation to en…
The Efficiency Frontier: A Unified Framework for Cost-Performance Optimization in LLM Context Management
Binqi Shen, Lier Jin, Hanyu Cai +2
Large language models (LLMs) increasingly rely on long-context processing, but expanding context windows introduces substantial computational and financial costs. Existing context…
Does Tone Change the Answer? Evaluating Prompt Politeness Effects on Modern LLMs: GPT, Gemini, and LLaMA
Hanyu Cai, Binqi Shen, Lier Jin +2
Prompt engineering has emerged as a critical factor influencing large language model (LLM) performance, yet the impact of pragmatic elements such as linguistic tone and politeness…
Rectify Evaluation Preference: Improving LLMs' Critique on Math Reasoning via Perplexity-aware Reinforcement Learning
Changyuan Tian, Zhicong Lu, Shuang Qian +8
To improve Multi-step Mathematical Reasoning (MsMR) of Large Language Models (LLMs), it is crucial to obtain scalable supervision from the corpus by automatically critiquing mistak…