11 papers
MultiTalk: Scaling Full-Duplex Speech Models to Long, Multi-Party, Bilingual Conversation
Ke Wang, Houxing Ren, Zimu Lu +4
End-to-end full-duplex speech models have brought open-source machine conversation closer to human-like interaction, yet existing systems remain limited in two intertwined dimensio…
Edit-Based Refinement for Parallel Masked Diffusion Language Models
Houxing Ren, Mingjie Zhan, Zimu Lu +5
Masked diffusion language models enable parallel token generation and offer improved decoding efficiency over autoregressive models. However, their performance degrades significant…
Towards Robust Real-World Spreadsheet Understanding with Multi-Agent Multi-Format Reasoning
Houxing Ren, Mingjie Zhan, Zimu Lu +4
Spreadsheets are central to real-world applications such as enterprise reporting, auditing, and scientific data management. Despite their ubiquity, existing large language model ba…
From Solver to Tutor: Evaluating the Pedagogical Intelligence of LLMs with KMP-Bench
Weikang Shi, Houxing Ren, Junting Pan +8
Large Language Models (LLMs) show significant potential in AI mathematical tutoring, yet current evaluations often rely on simplistic metrics or narrow pedagogical scenarios, faili…
FullStack-Agent: Enhancing Agentic Full-Stack Web Coding via Development-Oriented Testing and Repository Back-Translation
Zimu Lu, Houxing Ren, Yunqiao Yang +4
Assisting non-expert users to develop complex interactive websites has become a popular task for LLM-powered code agents. However, existing code agents tend to only generate fronte…
SlidesGen-Bench: Evaluating Slides Generation via Computational and Quantitative Metrics
Yunqiao Yang, Wenbo Li, Houxing Ren +6
The rapid evolution of Large Language Models (LLMs) has fostered diverse paradigms for automated slide generation, ranging from code-driven layouts to image-centric synthesis. Howe…
WebGen-Agent: Enhancing Interactive Website Generation with Multi-Level Feedback and Step-Level Reinforcement Learning
Zimu Lu, Houxing Ren, Yunqiao Yang +5
Agent systems powered by large language models (LLMs) have demonstrated impressive performance on repository-level code-generation tasks. However, for tasks such as website codebas…
Alignment with Fill-In-the-Middle for Enhancing Code Generation
Houxing Ren, Zimu Lu, Weikang Shi +7
The code generation capabilities of Large Language Models (LLMs) have advanced applications like tool invocation and problem-solving. However, improving performance in code-related…
Probability-Consistent Preference Optimization for Enhanced LLM Reasoning
Yunqiao Yang, Houxing Ren, Zimu Lu +6
Recent advances in preference optimization have demonstrated significant potential for improving mathematical reasoning capabilities in large language models (LLMs). While current…
MathCoder-VL: Bridging Vision and Code for Enhanced Multimodal Mathematical Reasoning
Ke Wang, Junting Pan, Linda Wei +8
Natural language image-caption datasets, widely used for training Large Multimodal Models, mainly focus on natural scenarios and overlook the intricate details of mathematical figu…
WebGen-Bench: Evaluating LLMs on Generating Interactive and Functional Websites from Scratch
Zimu Lu, Yunqiao Yang, Houxing Ren +7
LLM-based agents have demonstrated great potential in generating and managing code within complex codebases. In this paper, we introduce WebGen-Bench, a novel benchmark designed to…