Search Results for author:
Found 11 papers, 6 papers with code
TwT: Thinking without Tokens by Habitual Reasoning Distillation with Multi-Teachers' Guidance
To address this challenge, we propose TwT (Thinking without Tokens), a method that reduces inference-time costs through habitual reasoning distillation with multi-teachers' guidance, while maintaining high performance.
Uncovering the Impact of Chain-of-Thought Reasoning for Direct Preference Optimization: Lessons from Text-to-SQL
Direct Preference Optimization (DPO) has proven effective in complex reasoning tasks like math word problems and code generation.
Exploiting Task Relationships for Continual Learning Using Transferability-Aware Task Embeddings
Continual learning (CL) has been an essential topic in the contemporary application of deep neural networks, where catastrophic forgetting (CF) can impede a model's ability to acquire knowledge progressively.
TableMaster: A Recipe to Advance Table Understanding with Language Models
While current language models (LMs) excel at many text-based tasks, they still face challenges in table understanding due to the complex characteristics of tabular data, such as their structured nature.
CodeS: Towards Building Open-source Language Models for Text-to-SQL
To address the limitations, we introduce CodeS, a series of pre-trained language models with parameters ranging from 1B to 15B, specifically designed for the text-to-SQL task.
Enhancing Continuous Domain Adaptation with Multi-Path Transfer Curriculum
Specifically, we construct a transfer curriculum over the source and intermediate domains based on Wasserstein distance, motivated by theoretical analysis of CDA.
Refined Temporal Pyramidal Compression-and-Amplification Transformer for 3D Human Pose Estimation
Accurately estimating the 3D pose of humans in video sequences requires both accuracy and a well-structured architecture.
Ranked #10 on
3D Human Pose Estimation
on HumanEva-I
PoSynDA: Multi-Hypothesis Pose Synthesis Domain Adaptation for Robust 3D Human Pose Estimation
Typically, PoSynDA uses a diffusion-inspired structure to simulate 3D pose distribution in the target domain.
KeyPosS: Plug-and-Play Facial Landmark Detection through GPS-Inspired True-Range Multilateration
By spearheading the integration of Multilateration with facial analysis, KeyPosS marks a paradigm shift in facial landmark detection.
ChatPipe: Orchestrating Data Preparation Program by Optimizing Human-ChatGPT Interactions
Orchestrating a high-quality data preparation program is essential for successful machine learning (ML), but it is known to be time and effort consuming.
HDFormer: High-order Directed Transformer for 3D Human Pose Estimation
Human pose estimation is a challenging task due to its structured data sequence nature.
Ranked #89 on
3D Human Pose Estimation
on Human3.6M
