6 accepted papers
Fine-tuning large language models (LLMs) faces significant memory challenges due to the high cost of back-propagation. MeZO addresses this using zeroth-order (ZO) optimization, matching memory usage to inference but suffering from slow convergence due to varying curvatures across model parameters. T
Direct Preference Optimization (DPO) has emerged as an effective approach for mitigating hallucination in Multimodal Large Language Models (MLLMs). Although existing methods have achieved significant progress by utilizing vision-oriented contrastive objectives for enhancing MLLMs' attention to visua…
Long chain-of-thought (CoT) supervision has become a common strategy to enhance reasoning in language models. While effective for large models, we identify a phenomenon we call Long CoT Degradation, in which small language models (SLMs; ≤ 3B parameters) trained on limited long CoT data experience si
Recent advancements in pre-trained language models (PLMs) have demonstrated that these models possess some degree of syntactic awareness. To leverage this knowledge, we propose a novel chart-based method for extracting parse trees from masked language models (LMs) without the need to train separate…
While prompt tuning approaches have achieved competitive performance with high efficiency, we observe that they invariably employ the same initialization process, wherein the soft prompt is either randomly initialized or derived from an existing embedding vocabulary. In contrast to these conventiona…
The underlying mechanism of neural networks in capturing precise knowledge has been the subject of consistent research efforts. In this work, we propose a theoretical approach based on Neural Tangent Kernels (NTKs) to investigate such mechanisms. Specifically, considering the infinite network width,…