6 papers
MetaEvo: A Meta-Optimization Framework for Experience-Driven Agent Evolution
Bowen Ren, Heyan Huang, Yinghao Li +1
Large language models (LLMs) exhibit strong reasoning capabilities, yet most LLM-based agents are statically deployed and unable to improve through task interactions. Existing expe…
Scaling Speech Tokenizers with Diffusion Autoencoders
Yuancheng Wang, Zhenyu Tang, Yun Wang +9
Speech tokenizers are foundational to speech language models, yet existing approaches face two major challenges: (1) balancing trade-offs between encoding semantics for understandi…
Unveiling and Addressing Pseudo Forgetting in Large Language Models
Huashan Sun, Yizhe Yang, Yinghao Li +2
Although substantial efforts have been made to mitigate catastrophic forgetting in continual learning, the intrinsic mechanisms are not well understood. In this work, we demonstrat…
SRA-MCTS: Self-driven Reasoning Augmentation with Monte Carlo Tree Search for Code Generation
Bin Xu, Yiguan Lin, Yinghao Li +1
Large language models demonstrate exceptional performance in simple code generation tasks but still face challenges in tackling complex problems. These challenges may stem from ins…
Extrapolation Merging: Keep Improving With Extrapolation and Merging
Yiguan Lin, Bin Xu, Yinghao Li +1
Large Language Models (LLMs) require instruction fine-tuning to perform different downstream tasks. However, the instruction fine-tuning phase still demands significant computation…
PSST: A Benchmark for Evaluation-driven Text Public-Speaking Style Transfer
Huashan Sun, Yixiao Wu, Yuhao Ye +4
Language style is necessary for AI systems to understand and generate diverse human language accurately. However, previous text style transfer primarily focused on sentence-level d…