7 papers
APEX4: Efficient Pure W4A4 LLM Inference via Intra-SM Compute Rebalancing
Hong Guo, Nianhui Guo, Weixing Wang +3
W4A4 quantization promises full utilization of INT4 Tensor Cores, yet group dequantization overhead on CUDA Cores has driven existing systems to mixed-precision fallbacks. We prese…
Can Large Language Models Generalize Procedures Across Representations?
Fangru Lin, Valentin Hofmann, Xingchen Wan +4
Large language models (LLMs) are trained and tested extensively on symbolic representations such as code and graphs, yet real-world user tasks are often specified in natural langua…
Beyond Accuracy: Benchmarking Cross-Task Consistency in Unified Multimodal Models
Weixing Wang, Liudvikas Zekas, Anton Hackl +5
Unified Multimodal Models (uMMs) aim to support both visual understanding and visual generation within a shared representation. However, existing evaluation protocols assess these…
Inverse Reinforcement Learning with Dynamic Reward Scaling for LLM Alignment
Ruoxi Cheng, Haoxuan Ma, Weixin Wang +7
Alignment is vital for safely deploying large language models (LLMs). Existing techniques are either reward-based (training a reward model on preference pairs and optimizing with r…
Improving Neural Argumentative Stance Classification in Controversial Topics with Emotion-Lexicon Features
Mohammad Yeghaneh Abkenar, Weixing Wang, Manfred Stede +3
Argumentation mining comprises several subtasks, among which stance classification focuses on identifying the standpoint expressed in an argumentative text toward a specific target…
TraceDet: Hallucination Detection from the Decoding Trace of Diffusion Large Language Models
Shenxu Chang, Junchi Yu, Weixing Wang +4
Diffusion large language models (D-LLMs) have recently emerged as a promising alternative to auto-regressive LLMs (AR-LLMs). However, the hallucination problem in D-LLMs remains un…